Torrens and Recording - Land Title Assurance in The Computer Age

William Mitchell Law Review
Volume 18 | Issue 1 Article 4
1992
Torrens and Recording: Land Title Assurance in

the Computer Age
John L. McCormack
Follow this and additional works at: h7p://open.mitchellhamline.edu/wmlr
Recommended Citation
McCormack, John L. (1992) "Torrens and Recording: Land Title Assurance in the Computer Age," William Mitchell Law Review: Vol.
18: Iss. 1, Article 4.
Available at: h7p://open.mitchellhamline.edu/wmlr/vol18/iss1/4
6is Article is brought to you for free and open access by the Law Reviews
and Journals at Mitchell Hamline Open Access. It has been accepted for
inclusion in William Mitchell Law Review by an authorized administrator
of Mitchell Hamline Open Access. For more information, please contact
sean.felhofer@mitchellhamline.edu.
Mitchell Hamline School of Law
McCormack: Torrens and Recording: Land Title Assurance in the Computer Age
TORRENS AND RECORDING:

LAND TITLE ASSURANCE IN THE
COMPUTER AGE
JOHN L. MCCORMACKt
I. INTRODUCTION ..................................... 62
II. TITLE ASSURANCE IN THE UNITED STATES ........... 67
A. Development of Recording and Registration Systems in
the United States ................................ 67
1. The American Recording System ............... 67
2. The Torrens System of Title Registration........ 70
3. Computerization of Title Records .............. 73
B. Title Assurance: Goals and Methods ............... 74
1. Goals of Title Assurance.: Economy, Efficiency and
Security of Title ............................. 74
2. Methods of Title Assurance ................... 76
III. THE TORRENS TITLE REGISTRATION SYSTEM ........ 80
A. The Torrens System in Operation: Mirror, Curtain
and Indemnity .................................. 80
B. Initial Registration .............................. 83
C. Administrative Initial Registration of Possessory Titles 86
D. Torrens in Operation: Security of Title ............. 89
1. Caveats .................................... 91
2. Governmental Interests ....................... 92
a. FederalRight Exceptions .................. 92
b. State Governmental Interest Exceptions ...... 93
3. Private Special Interest Exceptions ............. 93
t Associate Professor of Law, Loyola University Chicago School of Law; Mem-

ber of the Illinois Bar; B.S. 1964, J.D. 1968, University of Wisconsin.
The author hereby acknowledges the valuable research assistance provided by Ms.
Karen L. Kaluza, Mr. Rocco S. DeFilippis, Ms. Elizabeth M. Krepps, and Ms. Nancy L.
Tuohy. Mr. Richard W. Edblom, Examiner of Titles, and Ms. Jan Witkowski, Regis-
tered Property Analyst, Hennepin County, Minnesota, were very helpful when the
author visited their offices in November 1990. Ms. Susan McIntosh, Real Property
Registration Branch, Province of Ontario Ministry of Consumer and Commercial Re-
lations, Toronto, Ontario, was of great assistance when the author visited Ministry
offices in July 1991. Professors Jane Hoffman Locke, Richard A. Michael and Patrick
M. McFadden, as well as Mr. Edblom, made valuable comments on earlier drafts of
this article. The author is solely responsible for any errors. A summer research grant
from Loyola University Chicago provided research support.
Published by Mitchell Hamline Open Access, 1992 1

William Mitchell Law Review, Vol. 18, Iss. 1 [1992], Art. 4 [Vol. 18
WILLIAM MITCHELL LAW REVIEW
4. Possessory Interest Exceptions .................. 94

5. Due Process, Notice and Equitable Exceptions ... 95
6. Error Exceptions ............................. 96
7. Encroachments .............................. 96
8. Traditionally Non-Title-Related Exceptions ..... 97
9. Conclusion: Security of Torrens Titles .......... 98
E. User Difficulties with Torrens ..................... 99
F. Operation and Administration of Torrens: Economy
and Efficiency ? .................................. 103
IV. COMPUTERIZED RECORDING ........................ 115
A. Registration and Recording: Land Title Assurancefor
the Computer Age ............................... 115
B. Basic Components ............................... 116
C. Registration or Recording? ....................... 121
D. Mirror, Curtain and Indemnity in Computerized
Recording Systems ............................... 123
1. M irror ..................................... 124
2. Curtain .................................... 125
3. Indem nity ................................... 127
V. CONCLUSION ....................................... 128
I. INTRODUCTION
The weaknesses of the recording systems' used in the

United States are well known. Over the years, commentators
have proposed replacing recording systems with the Torrens
system of land title registration. Some proponents of Torrens
point to the apparent success of registration systems in foreign
countries, particularly Canada and the United Kingdom, as
1. The term "recording" will be used to refer to land title record systems where
the government simply acts as a custodian of land title data. "Title registration" will
refer to systems where the government evaluates data, usually as it comes in, and
where the record maintained is intended to be a generally conclusive governmental
statement concerning ownership and title. The term "Torrens" will refer specifically
to the title registration system adopted in Australia in 1857 and systems found else-
where in the world which are directly based on the Australian model.
Some commentators use "Torrens" to refer to systems, actual or proposed,
which may include some aspects of the Australian model without intending to refer to
such model in its entirety. Of course, classifications are rarely perfect and some sys-
tems do not neatly fit into either the "recording" or "registration" categories as de-
fined above. Indeed, title registration systems usually store some data for notice
purposes without affirming that the data represent valid interests, while some record-
ing systems give affirmations that certain data stored in the database represent legally
binding interests or events.
http://open.mitchellhamline.edu/wmlr/vol18/iss1/4 2
1992] McCormack: TorrensTORRENS
and Recording:
AND Land Title Assurance in the Computer Age
RECORDING
proof that title registration could work in the United States.2

One commentator observed:
It is a baffling fact that the United States is rapidly becoming
virtually the only country in the world whose land title sys-
tem is not founded upon Torrens-type principles. The
writer finds it incredible that a system which seems to work
quite well almost everywhere else cannot be satisfactorily
adapted to the United States.'
The published commentary on the Torrens system in the
American legal journals is overwhelmingly favorable. 4 Despite
this, attempts to implement the Torrens system in the United
States have almost completely failed. 5 Torrens advocates attri-
2. See Ted J. Fiflis, Security and Economy in Land Transactions: Some Suggestions from
Scotland and England, 20 HASTINGS L.J. 171, 177-200 (1968) [hereinafter Fiflis, Security
and Economy].
3. Walter E. Barnett, Marketable Title Acts-Panacea or Pandemonium?, 53 CORNELL
L. REV. 45, 93-94 (1967). The United States is hardly alone in using recording sys-
tems, and the movement toward registration is proceeding at a glacial pace. See Fiflis,
Security and Economy, supra note 2, at 177-200.
4. See generally Barry Goldner, Comment, The Torrens System of Title Registration: A
New ProposalforEffective Implementation, 29 UCLA L. REV. 661, 665-67 (1982) (contain-
ing an excellent bibliography current to 1982).
Much of the American commentary, both pro and con, on Torrens is flawed be-
cause it is advocacy, rather than balanced scholarship. It also frequently suffers be-
cause of the authors' unfamiliarity with the subject. Anyone seriously interested in
Torrens should study the Australian and Canadian commentary.
5. One respected scholar stated that, with the exception of the United States, he
"knows of no jurisdiction in which title registration can be said to have failed." 1
THOMAS W. MAPP, TORRENS' ELUSIVE TITLE 3 (Alberta Law Review Book Series,
1978).
Although it is commonly believed that title registration has failed only in the
United States, registration has actually failed in a number of places both in and
outside of the United States. For example, Nova Scotia, Canada, abandoned the Tor-
rens system which was established there around the turn of the century. R.C.B. Risk,
The Records of Title to Land: A Plea for Reform, 21 U. TORONTO L.J. 465, 470 (1971).
Recently, title registration has been revived in Nova Scotia. The maritime provinces
(New Brunswick, Nova Scotia and Prince Edward Island) have been using recording
systems for over 200 years. However, after much discussion and study, New Bruns-
wick and Nova Scotia recently enacted title registration statutes. Donald M. Lamont,
Land Registration Systems, in SURVEY LAw IN CANADA 3.103, at 104 (1989). This legis-
lation grew out of an ambitious program to improve and computerize public land
records in the maritimes. See Cyril Carlin, Computerization of Land Records in the Mari-
time Provinces of Canada, 43 U. CIN. L. REV. 487 (1974). Torrens was adopted in Brazil
in 1890 but is not used much there today. ALEJANDRO M. GARRO, THE LOUISIANA
PUBLIC RECORDS DOCTRINE AND THE CIVIL LAW TRADITION 36, at 81 n.18 (1989).
Torrens also failed in Spain and other parts of Latin America. Id. 36, at 81.
Most of the countries using Torrens today were British colonies when Torrens
was adopted and had no established public land title record systems at that time. To
date, no title registration system has successfully replaced an entrenched recording

William Mitchell Law Review, Vol. 18, Iss. 1 [1992], Art. 4
WILLIAM MITCHELL LAW REVIEW [Vol. 18
bute this failure to the high cost of initial registration6 and the
opposition to the Torrens system from the title assurance in-
dustry.7 While this explanation contains some truth, it is
somewhat simplistic and certainly incomplete.
The most significant cause of the failure of Torrens in the
United States was the inertia of entrenched recording systems.
Major contributing causes included the high cost of initial re-
gistration and the cost and difficulty involved in the continuing
administration of the Torrens system. 8 This latter cause of the
failure of Torrens has been very little appreciated. 9 It is
doubtful whether opposition from the title assurance industry
played a significant role.
The goal of the Torrens system is to establish and maintain a
governmental record which mirrors the precise, current state
of title. This goal has two consequences. First, creating a
"mirror" makes initial title registration expensive. Second, ad-
ministering such a system is relatively expensive and burden-
some for government.' Without adequate funding and
competent people to accurately and promptly evaluate title
system in any major independent jurisdiction in the world. See id. Currently, projects
are underway to replace entrenched recording systems with registration in Scotland,
Ireland (Northern and the Republic), Ontario and the maritime provinces of Canada.
6. Under almost all American Torrens acts, a judicial proceeding is required to
initially register a title. Torrens advocates argue that the cost of initial registration
could be reduced through initial administrative registration of possessory titles that
would be inconclusive for a period of time. See Goldner, supra note 4, at 690.
7. See Ted J. Fiflis, Land Transfer Improvement: The Basic Facts and Two Hypotheses for
Reform, 38 U. COLO. L. REV. 431, 444-50 (1966) [hereinafter Fiflis, Land Transfer
Improvement].
8. Under Torrens, every instrument presented for registration must be evalu-
ated prior to its registration. In contrast, evaluation of instruments under recording
systems only occurs when the title is searched and examined. The latter method of
evaluating instruments in groups is arguably more efficient than evaluating them in-
dividually. Evaluation of a subsequent transaction may make evaluation of prior ones
unnecessary. In addition, the passage of time may make evaluation unnecessary.
Commentators favoring Torrens often emphasize the wasteful repetition of data eval-
uation under recording but ignore the inefficiency in continuous data evaluation
under Torrens.
9. See infra text accompanying notes 213-24.
10. Report of the Blue Ribbon Committee Prepared at the Request of Carol
Moseley Braun, Cook County Recorder of Deeds/Registrar of Titles 51 (June 13,
1989) (unpublished report; on file with author) [hereinafter Cook County Blue Rib-
bon Report]. Registration systems usually require subsidies because the costs of op-
eration generally exceed the income from fees and other sources. However, the
Hennepin County, Minnesota, Torrens system is currently at least self-sufficient.
Hennepin County has supported its operations through economies (such as avoiding
unnecessary issuance of certificates, electing not to check documents for forgery) and
19921
McCormack: TorrensTORRENS
and Recording: Land Title Assurance in the Computer Age
AND RECORDING
data, the registration office will be unable to adequately admin-

ister the system.''
The architects of the American Torrens acts apparently gave
little thought to whether county governments would be
equipped or inclined to finance and implement the system.
Perhaps it was believed that the inherent superiority of Tor-
rens would inexorably lead to its implementation. The archi-
tects may have believed that after state government authorized
Torrens, demand from the public for replacement of the re-
cording system would assure implementation. The authors of
Torrens legislation failed to appreciate that demand from the
public would not be forthcomming in sufficient strength be-
cause of ignorance and acceptance of the status quo. Because
of the costs and difficulties involved in administration of Tor-
rens, local government was not likely to be an impetus for its
implementation. Ordering county governments to implement
Torrens was not a realistic alternative absent fundamental
changes in the pattern of state and local government. The
elected county recorders were not directly responsible to any-
one in a hierarchy above. County government legislators were
directly accountable to practically no one but their
constituents.
Most commentary on the comparative merits of title registra-
tion and recording systems focuses on whether registration
should replace recording. It is generally believed, even by
those who think registration is superior to recording, that title
registration has no realistic possibility of being implemented
through fees which produce adequate revenues. Edblom-Witkowski Interview, infra

note 72.
11. The Torrens system in Cook County, Illinois, failed after 91 years precisely
because the government failed to provide the resources necessary to make the system
reasonably efficient. See FRED I. FEINSTEIN, The Torens System, in 2 BASIC REAL ESTATE
PRACTICE 1-7, 15-3 (Ill. Inst. for Continuing Legal Educ., 1988 & 1990 Supp.). Over
the years, Illinois has developed the following improvements to its Torrens system:
1) creating tax lien files for federal and state tax liens and requiring the Registrar to
search for and memorialize filed liens against a registered title whenever a memorial
is entered on the original certificate for such title; 2) replacing lost certificates by
affidavits without resort to subsequent proceedings; 3) using title indemnity bonds to
induce the Registrar to remove memorials of encumbrances which are probably no
longer valid without resort to subsequent proceedings; 4) administratively processing
the devolution of decedents' interests; 5) requiring judgment lien creditors to memo-
rialize their judgments on original title certificates; and 6) backing up the assurance
fund with the assets and credit of the county. See ILL. ANN. STAT. ch. 30, 84, 102,
109.1, 114-15, 122, & 139.1 (Smith-Hurd 1969 & Supp. 1991).

William Mitchell
WILLIAM Law Review,
MITCHELLVol. 18,LAW
Iss. 1 [1992],
REVIEW
Art. 4 [Vol. 18
throughout the United States in the foreseeable future.' 2

While implementing title registration nationwide may not be
possible, experience with it may be used to improve recording
systems.
Those seeking to improve land title assurance systems must
also take into account developments in computer technology.
Computerization can reduce the comparative attractiveness of
Torrens over recording. 13 While title registration systems
have their own weaknesses compared to recording, Torrens is
clearly superior from the title examiner's perspective in its
consolidation of most relevant information in a certificate or
register. Consequently, the data management and retrieval
characteristics of manual Torrens systems are clearly superior
to those of manual recording systems. 4 Computerization of
Torrens does relatively little to improve these already good
data management and retrieval characteristics. In contrast,
computerization of recording systems can do much to improve
their data management and retrieval capability by replacing
slow and cumbersome manual methods of assembling relevant
title data with rapid electronic means.
It is quixotic to advocate the implementation of title registra-
tion throughout the United States. The present county-by-
county system is too well-entrenched to be summarily replaced
12. See JOHN E. CRIBBET & CORWIN W.JOHNSON, PRINCIPLES OF THE LAW OF PROP-
ERTY 352 (3d ed. 1989).
13. But see Martin Lobel, A Proposalfor a Title Registration System for Realty, 11 U.
RICH. L. REV. 501, 511 (1977). Lobel criticizes computerization proposals as "ex-
pressing the naive notion that technology and automation are the keys to the solution
of societal problems .. " Id, However, when a large part of the problem is data
input, storage and retrieval, it is hardly naive to expect that computers may provide
at least a partial solution. Lobel also states that "proponents of computerization
tend to overlook the fact that the start-up and capital costs are often prohibitive." Id.
Concern about start-up costs seems odd coming from an advocate of title registra-
tion. The cost of initially registering all titles in the District of Columbia, as pro-
posed by Lobel, would have been very high.
When Lobel published his article in 1977, the microcomputer revolution was
just getting underway. Powerful computer technology is now affordable. Start-up
costs can be drastically lowered by making the computerized operations prospective
only, as was done in some computerized public records and private title plants. An-
other alternative is to backload data entered since the property was last transferred
or during some fixed period into the past, such as thirty or forty years. Backloading
makes a computerized system more useful to title searchers in the early stages of
operations.
14. Colin D. Hadfield, Computerizationof Land Title Records, 43 U. CIN. L. REV. 513,
514 (1974).
19921
McCormack: Torrens TORRENS
and Recording:
RECORDING
by special land registration districts administered by qualified

civil servants under direct state supervision. The replacement
of private sector title assurance is unwarranted, absent a clear
showing that doing so would result in a substantial savings to
the public. A better aim now is to improve existing recording
systems in order to promote economy, efficiency and security
of title in real estate transactions by applying computer tech-
nology and experience with title registration in the United
States and elsewhere. 5
II. TITLE ASSURANCE IN THE UNITED STATES
A. Development of Recording and Registration Systems in the United

States
1. The American Recording System 16
Assurance for most real estate titles in the United States is
based on recording systems of the type first used in the Massa-
chusetts Bay Colony in the 1600s.'7 Today, the American re-
cording system usually operates under state law at the county
level of government. 8 Under this system, documents which
may affect title to real estate are presented to government of-
fices for recordation. Recording perfects legal priority over
possible conflicting interests, thereby protecting the holder of
an instrument against the possible loss of ownership or prior-
ity. In the absence of such perfection, title may be lost to a
subsequent transferee who qualifies for the protection of the
recording system.' 9
Recording is typically not a prerequisite to legal validity. Ex-
ecuted deeds and other instruments may create interests in
property even if they are not recorded. In addition, recording
a void instrument does not normally make it effective, although
15. See Dale A. Whitman, Optimizing Land Title Assurance Systems, 42 GEO. WASH. L.
REV. 40, 65 (1973).
16. Many good sources have been published discussing the American recording
system, its weaknesses, proposed improvements and improvements already
implemented. See, e.g., CRIBBET &JOHNSON, supra note 10, at 346; John E. Cribbet,
Conveyancing Reform, 35 N.Y.U. L. REV. 1291 (1960).
17. George L. Haskins, The Beginnings of the Recording System in Massachusetts, 21
B.U. L. REV. 281, 282 (1941).
18. In a few places, charter cities, towns or townships maintain and administer
public land title records. U.S. DEP'T OF HUD, LAND TITLE RECORDATION PRACTICES:
A STATE-OF-THE-ART STUDY 12 (1980) [hereinafter HUD, STATE-OF-THE-ART STUDY].
19. CRIBBET &JOHNSON, supra note 12, at 307.

WILLIAM MITCHELL LA W REVIEW [Vol. 18
recording may raise a presumption of validity. 20 Furthermore,

the acceptance of an instrument for recordation does not usu-
ally reflect a governmental judgment that the instrument is le-
gally effective. Instead, the government is merely a depository
of copies of the instruments so that parties wishing to evaluate
recorded documents may have access to them. In this respect,
recording facilitates real estate transfers by giving prospective
transferees information which is relevant to the determination
of ownership. 2 '
Generally, recording systems use either grantor-grantee or
tract indexes to locate recorded documents. Locating all of
the relevant recorded documents can be difficult or even im-
possible using the available indexes. The oldest and most
common type of index to the copies of recorded instruments is
the grantor-grantee type. In this type of index, instruments
are indexed alphabetically according to the grantors' and
grantees' surnames. The grantee index is used to reach back
into time to establish the chain of owners. The grantor index
is used to find adverse recorded conveyances made by or
through each owner during the time that the owner was the
apparent, actual or record owner of the interest being
searched. The grantor-grantee type of index is relatively easy
and inexpensive for government to administer, but it is nor-
22
mally difficult to use.
The tract type of index is much easier to use but is more
difficult and expensive for government to maintain.23 In states
without official tract indexes, government or private title com-
20. 1 RUFFORD G. PATRON & CARROLL G. PATON, PATRON ON LAND TITLES 20

(2d ed. 1957). This presumption is usually rebuttable. Id.
21. MAPP, supra note 5, at 49.
22. Searching a title with a grantor-grantee index can be physically cumbersome
and time-consuming because a great number of index books may have to be con-
sulted and reconsulted. In addition, where a past transfer of title does not appear in
the grantee index, the title searcher may have to guess how ownership may have
passed to an owner in order to reach further back in time where additional transac-
tions may be recorded. If a particular transaction does not appear in the grantee
index, the searcher is limited to the process of trial and error and may or may not be
able to discover how ownership passed to a particular owner.
23. Recording office employees must be able to identify the proper segment of
the index in which to reference instruments, usually from the legal descriptions ap-
pearing on each instrument. This process is time-consuming, more costly, and re-
quires a higher level of expertise than is the case with a grantor-grantee index.
Consequently, official tract indexes are found in relatively few states. See ROGER A.
CUNNINGHAM ET AL., THE LAW OF PROPERTY, 11.11, at 797 (1984).
1992] TORRENS AND RECORDING
panies sometimes maintain unofficial tract indexes. Tract in-

dexes organize instruments according to the property they
affect. Instruments affecting each segment of land are indexed
on a page or set of pages for that parcel. Once the proper
portion of the index has been located, searching title is rela-
tively simple.2 4
The recording system falls short of providing conclusive se-
curity of real estate ownership. 2 5 It does not ensure that the
actual state of ownership and the record of ownership are the
same or even similar. A recorded, apparently valid transaction
may be void or defective. Unrecorded interests that are dis-
coverable by physical inspections or inquiries may be valid
under the doctrines of constructive notice from possession and
inquiry notice. Furthermore, some unrecorded interests may
be valid even if they are not discoverable by such inspections
or inquiries.2 6
24. Title can be searched using the same segment of the index until the point in
the past where the index was reorganized. Where this occurs, there is usually a refer-
ence to the older portion of the index where instruments affecting the subject prop-
erty were indexed prior to the reorganization. Where a past title transfer was not
indexed, the searcher does not run into the dead end which may be encountered
using the grantor-grantee index.
25. For another good discussion of the deficiencies in the recording system, see
Harry M. Cross, Weaknesses of the Present Recording System, 47 IOWA L. REV. 245 (1962).
26. A recorded instrument may be valid on its face but still be void, voidable or
defective. The recording system does not protect a transferee who fails to receive an
interest because it is based on a previously recorded void or defective instrument or
transaction.
In addition, certain interests or claims may be valid against a particular title even
though nothing appears in the record concerning them. See DEP'T OF HUD, LAND
TITLE RECORDATION SYSTEMS: LEGAL CONSTRAINTS AND REFORMS IV-I (1979) [herein-
after HUD, LEGAL CONSTRAINTS AND REFORMS]. Short term leases are exempt from
recordation. See 4 AMERICAN LAW OF PROPERTY 17.8 n.10 (A. James Casner ed.,
1952). Usually, possession by the lessee will give notice of the lease. Certain claims
or interests are outside of the recording system because they are not created or trans-
ferred by written instruments: adverse possession and prescriptive easement claims,
implied easements, marital or homestead rights and unfiled mechanics' liens. See gen-
erally CUNNINGHAM, supra note 23, at 780. Possession will usually give notice of an
adverse possession or prescriptive easement claim. A purchaser is also deemed to
have notice of facts which would have been uncovered through a physical inspection
of the property. This gives rise to the doctrine of constructive notice based on pos-
session. Under this doctrine, an unrecorded claim may have priority over a subse-
quent claim if a physical inspection of the real estate would give notice of it.
Under the Supremacy Clause of the U.S. Constitution, U.S. CONST. art. VI, cl. 2,
claims, rights and other interests based on federal law are not subject to state record-
ing laws unless the United States consents. The difficulty caused by these interests
has been ameliorated because the United States generally requires its agencies to file
or record (but not to register) federal real estate claims and has consented to their

2. The Torrens System of Title Registration

Commentators have exhaustively discussed the weaknesses
of the American recording system as a source of land title as-
surance. 27 Many reforms of the system have been proposed
and a few have even been implemented. 2 8 One of the most
radical reform proposals is to replace the recording system
with the Torrens system of title registration.2 1 Unlike record-
ing, title registration does not usually create or transfer a legal
interest until government itself makes a conclusive assessment
of the current state of the title.
While title registration has been used in continental Europe
since the early Middle Ages,3 0 modern European title registra-
tion systems were not established until the 1800s." t During
subordination in the event of a failure to file or record. See, e.g., The Federal Tax
Lien Act, 26 U.S.C. 6323 (1989); Davis v. United States, 705 F. Supp. 446, 453
(C.D. Ill. 1989).
In addition to matters which may be valid without any reference at all to them in
public records, other possible liens, claims, encumbrances or other interests may be
binding on subsequent purchasers even though it might be extremely difficult, if not
practically impossible, to find information about them in public records. See HUD,
LEGAL CONSTRAIrs AND REFORMS, supra, at 111-39. These include bankruptcy trans-
fers, exercise of rights of eminent domain, matters referred to in recorded instru-
ments, interests recorded before or after the time the record shows their transferors
to have ownership and judgment liens filed alphabetically by debtors' surnames, es-
pecially where the names are common. See 11 U.S.C. 549(c) (1990). Under the
doctrine of idem sonans, a party may be deemed to have notice of information indexed
under incorrectly spelled names which sound similar to the owner's name. See HUD,
LEGAL CONSTRAINTS AND REFORMS, supra, at 5, 11-26.
27. Much of this commentary exaggerates the inefficiencies and weaknesses of
the recording system. See, e.g., Goldner, supra note 4, at 665-67. For a comprehen-
sive and generally balanced discussion, see generally HUD, LEGAL CONSTRAINTS AND
REFORMS, supra note 24.
28. The principal reforms adopted include Marketable Title acts and official tract
(parcel) indexes.
29. See Martin Lobel, A Proposalfor a Title Registration System for Realty, 11 U. RICH.
L. REV. 501 (1977) (proposing the adoption of a Torrens system for the District of
Columbia).
30. SIR ERNEST DOWSON & V.L.O. SHEPPARD, LAND REGISTRATION, COLONIAL RE-
SEARCH PUB. No. 13, 173 (2d ed. 1964); RUDOLF HUEBNER, A HISTORY OF GERMANIC
PRIVATE LAW 218-19 (Francis S. Philbrick trans., reprint 1968) (1918).
31. DowsON & SHEPPARD, supra note 30, at 63, 180, 187; HUEBNER, SUpra note 30,
at 252. Land title record systems similar in basic concept to the American recording
system were adopted in Europe many years ago and are still used today. European
countries using recording include Belgium, France, Greece, the Republic of Ireland,
Northern Ireland, the Netherlands and Scotland. Ireland, Northern Ireland and
Scotland currently are converting very slowly to the English form of title registration.
Conveyancing in France really depends on notaries for title assurance. Notaries in
France in some respects play a role in conveyancing similar to that of title insurance
19921 TORRENS AND RECORDING
the 1800s, England considered adopting a recording system

but instead adopted a registration system in 1862.32 To a lim-
ited extent, England used some antecedents of the American
recording system, but a comprehensive land title records sys-
tem did not exist in England until this century. After substan-
tial modifications and a number of false starts," registration
was finally implemented throughout England and Wales after
companies in the United States. European countries using title registration include
the United Kingdom, the Republic of Ireland, Germany, Austria, Hungary, Poland,
Czechoslovakia, Yugoslavia, Denmark, Norway, Sweden and Switzerland. JOHN H.
CRABB, THE CONSTITUTION OF BELGIUM AND THE BELGIAN CIVIL CODE 347 (1982);
DOWSON & SHEPPARD, supra note 28, at 98; F.H. LAWSON ET AL., AMOS AND WALTON'S
INTRODUCTION TO FRENCH LAW 107-08 (3d ed. 1967); RICHARD R. POWELL, THE NEW
YORK LAW SOCIETY, REGISTRATION OF THE TITLE TO LAND IN THE STATE OF NEW YORK
287-93 (1938); DAVID M. WALKER, THE SCOTrISH LEGAL SYSTEM 119 (4th ed. 1976);
Athanassios N. Yiannopoulos, Property, in INTRODUCTION TO GREEK LAW 110-11 (Kon-
stantinos D. Kerameus & Phaedon J. Kozyris eds., 1988); HUD, STATE-OF-THE-ART
STUDY, supra note 18, at 23-24; John E. Cribbet, Some Reflections on the Law of Land-A
View from Scandinavia, 62 Nw. U. L. REV. 277, 282-83 (1967); Fiflis, Security and Econ-
omy, supra note 2, at 177-20; FRED I. FEINSTEIN, The Torrens System, in 2 BASIC REAL
ESTATE PRACTICE 1-7, 15-3 (Illinois Institute for Continuing Legal Education 1988 &
Supp. 1990); W.M. Kleijn, The Law of Property, in INTRODUCTION TO DUTCH LAW FOR
FOREIGN LAWYERS 86-87 (D.C. Fokkema et al., eds. 1978); M. Lewis & Martin H.
Jones, Scotland, in 3 DOING BUSINESS IN THE UNITED KINGDOM 53.05[1][g] (MB
1990); H. Wallace, Northern Ireland, in 3 DOING BUSINESS IN THE UNITED KINGDOM,
supra, 54.04[7]; Ussher, Land and Environmental Laws, in DOING BUSINESS IN IRELAND
8.01[3] (MB 1990). Although compulsory registration was authorized in Northern
Ireland in 1970, no compulsory registration area had been designated as of the late
1980s. In 1964, compulsory registration was authorized in the Republic of Ireland.
In Meath, Carlow and Laois Counties, registration is compulsory upon transfer by
sale or long term lease. Because of the condition of the records maintained under
the Irish recording system, searching titles in it is difficult, inconclusive and expen-
sive. Ussher, supra, 8.01[3]. The only substantial survey of foreign land title record
practices published in English since Dowson and Sheppard's study, supra note 28,
was conducted by Alejandro Garro of Columbia University. GARRO, supra note 5.
Garro's survey does not purport to be comprehensive. He discusses American and
French recording, Torrens and German title registration and, to a limited extent,
Latin American recording systems. Very little about Asian land title record practices
has been published in English. See Thomas H. Clarke, Mortgage and Title Assurance in
Asia, 11 INT'L Bus. LAW. 127 (1983). Torrens was established in Brazil in 1890 but
did not take hold. It is not used much there today. It also failed in other parts of
Latin America and Spain. GARRO, supra note 5, at 81 n.18; Francisco V. Igartua, The
Public Registry of Property in Mexico, 11 MIAMI L.Q. 457 (1957). Puerto Rico uses nota-
ries, registration and title insurance. U.S. DEP'T OF HUD AND VA, REPORT ON MORT-
GAGE SETTLEMENT COSTS 1213 (Supp. 1972) [hereinafter HUD/VA REPORT]. Land
title assurance in Italy appears to be based on notaries. Paul B. Rava, Italian Procedure
in Real Estate Conveyancing, 1964 A.B.A. PROCEEDINGS OF THE SECTION OF INTERNA-
TIONAL AND COMPARATIVE LAW (1965).
32. DowsON & SHEPPARD, supra note 30, at 39; C. Dent Bostick, Land Title Regis-
tration: An English Solution to an American Problem, 63 IND. LJ. 55, 74-75 (1987).
33. Land Transfer Act, 1875, 38 & 39 Vict., ch. 87 (Eng.).

enactment of the Land Registration Act of 1925. 34 In the late

1850s, Sir Robert Richard Torrens invented and implemented
the Torrens system of title registration in Australia.3 5
Sir Robert based his system on the English method of regis-
tering ships. Each ship was assigned a page in the registry
- where the name and description of the ship appeared, along
with the name of the owner and a statement of the liens and
encumbrances against it. The owner received a duplicate of
the page as a certificate of title and proof of ownership. Upon
sale of the ship, the instrument of transfer and the certificate
were sent to the registry office and a new page was prepared to
show the transfer of ownership.
Sir Robert reasoned that a similar system could be used to
register ownership of real estate and drafted legislation to ac-
complish that result. He promised that the new system would
have four "grand characteristics": certainty, economy, simplic-
ity and facility.3 6 In 1857, the legislation was enacted in South
Australia. Enactment of similar legislation in other British ter-
ritories soon followed. 7
Torrens is one of five general types of title registration sys-
tems used in the world3 8 and is the only type used in the
United States. In 1895, Illinois became the first American state
to enact Torrens legislation. 9 Within a few decades, Torrens
34. Land Registration Act, 1925, 15 & 16 Geo. 5, ch. 21 (Eng.). The English
program of compulsory registration is approaching completion of the mammoth task
of registering every parcel in England and Wales. Bostick, supra note 32, at 74-77;
Fiflis, Security and Economy, supra note 2, at 201-06.
35. A.G. Lang, Computerised Land Tile and Land Information, 10 MONASH U. L. REV.
196, 197 (1984).
36. Id.
37. British Honduras (Belize), 1859; Vancouver Island (British Columbia), 1860;
Tasmania, 1862; New South Wales, 1862; Ireland, 1865; New Zealand, 1870; Wales,
1875;Jamaica, 1888; Nova Scotia, 1904; and Uganda, 1908. BLAIR C. SHIcK & IRVING
H. PLOTKIN, TORRENS IN THE UNITED STATES 17 (1978); HUD, STATE-OF-THE-ART
STUDY, supra note 18, at 23.
38. DoWSON & SHEPPARD, supra note 30, at 98. The other four are English, Ger-
man, Swiss and Ottoman. Id.
39. The system was approved by the voters in Cook County on November 5,
1895. The statute was held unconstitutional in the following year. People ex rel. Kern
v. Chase, 46 N.E. 454 (Ill. 1896). A second act was adopted and approved by the
voters in 1897. This statute was held constitutional. People ex rel. Deneen v. Simon,
52 N.E. 910 (Il1. 1898). The Torrens office in Cook County opened on March 1,
1899, and has been in continuous operation ever since. InJanuary 1991, legislation
was signed which abolished Torrens in Illinois, effective January 1, 1992. 1990 I11.
Legis. Serv. 2959 (West); see also David Heckelman, Bill Abolishes Torrens Title Registra-
tion, CHI. L. BULL., Jan. 15, 1991, at 1.
was adopted by nineteen other states.4" Since then, the legisla-

tion has lapsed or has been repealed in nine of them.4 ' The
Torrens system is used to a substantial extent today in only five
states: Hawaii, Illinois, Massachusetts, Minnesota and Ohio.42
In Hawaii and Massachusetts, Torrens is used statewide. In
the other states, use is limited to a few localities: Illinois (Cook
County only), Minnesota (Hennepin and Ramsey Counties
with minimal registrations elsewhere) and Ohio (Hamilton
County with minimal registrations elsewhere). In no state or
locality are a majority of parcels registered under Torrens.
The highest incidence of use is probably in Hawaii where
nearly forty-five percent of all parcels are registered.43
3. Computerization of Title Records

The computerization of title records at title insurance com-
panies commenced in earnest in the 1960s.44 Today, title
searches in private title plants and public records are often
done on computer databases. These unofficial computer sys-
tems bear little resemblance to the manual recording system,
with its cumbersome indexes, or to the Torrens system, with
its archaic manual certificate registers.
In 1984, Ontario, Canada, became one of the first North
American jurisdictions to authorize an official computerized
system for both its recording and title registration systems.4 5
40. U.S. DEP'T OF HUD, LAND TITLE REGISTRATION SYSTEMS, PROPOSALS FOR IM-
PROVEMENT, 111-4 (1979) [hereinafter HUD, LAND TITLE REGISTRATION]; see also
HUD/VA REPORT, supra note 31, at II-A-4.
41. HUD/VA REPORT, supra note 31, at II-A-4, II-A-9; HUD, LAND TITLE REGIS-
TRATION, supra note 40, at 111-4. The National Conference of Commissioners on Uni-
form State Laws adopted the Uniform Land Registration Act in 1916. This act was
withdrawn by the commissioners as obsolete in 1934. HUD, LAND TITLE REGISTRA-
TION, supra note 40, at 111-6.
42. See HUD, LAND TITLE REGISTRATION, supra note 38, at 111-4, 111-5.
43. Id. at 111-10. In Hawaii, Torrens did not have to displace a well entrenched
recording system.
44. Fiflis, Security and Economy, supra note 2, at 172-73.
45. In 1984, Ontario, which uses both recording and title registration, adopted
an official computerized land title records system called "POLARIS" for both its re-
cording and title registration systems. Land Registration Reform Act, R.S.O. ch. 32
(1984) (Can.). Ontario recently decided to convert all of the Province to the land
titles (title registration) system as part of implementing POLARIS. In February
1991, the Provincial government announced the formation of a partnership called
Tenant Land Information Services Inc. between the Province and a private sector
organization, Real Data Ontario, which will implement POLARIS and play a large
role in administering it in the future. Interview with John Dalgliesh, Senior Adviser,

The Ontario system is now operational and in the process of

full implementation. Data storage and retrieval is completely
electronic. The future will produce more of the same. Data
retrieval through computers and land parcel identifiers will re-
place the current manual index methods of retrieving data, es-
pecially in populous urban jurisdictions.
B. Title Assurance: Goals and Methods

1. Goals of Title Assurance: Economy, Efficiency and Security of
Title
A basic objective of a land title assurance system is to enable
real estate interests to "move freely and easily in commerce. '"46
Dean John Cribbet has identified three principles which should
be observed to achieve this objective:
First, the system must give adequate security for land titles.
Unless the purchaser or mortgagee can be assured that his
investment is sound, the particular method fails, whatever
other virtues it may possess. Second, it must provide speed
in the determination of title status so that the transaction
can be closed with a minimum amount of cliff-hanging.
Third, the method must be relatively inexpensive so that a dis-
proportionate amount of the purchaser's dollar is not chan-
neled into title service. 4 7
The first principle, security of titles, refers to fulfillment of
expectations about ownership and use of real estate. A system
of title assurance should generate reliable expectations in
transferees and owners about the legal rights and liabilities as-
sociated with the ownership and use of real estate.4 8 Transfer-
ees and owners should have assurance .that (1) they will have
the legal right to possess or use all of their interests in the sub-
ject real estate for their intended purposes, (2) there are no
undisclosed financial liabilities attached to its ownership, (3)
Operations, for the Strategic Alliance Liason, Ministry of Consumer and Commercial
Relations, Province of Ontario, in Toronto, Can. (July 25, 1991) [hereinafter Dal-
gliesh Interview]. In the Ontario system, data will be retrievable using permanent
parcel identifiers, persons' names, tax assessment numbers, street addresses or num-
bers assigned recorded or registered documents. John R. Dow, Property and Real Es-
tate Law, in 2 DOING BUSINESS IN CANADA 9.06[1][d] (MB 1990).
46. CRIBBET &JOHNSON, supra note 12, at 346; see also Cribbet, supra note 16, at
1291.
47. CRIBBET &JOHNSON, supra note 12, at 347-48 (emphasis in original).
48. Stephen Cretney, English Conveyancing Practice, in HUD/VA REPORT, supra note
31, at 1266 app.
the ownership of the interest will remain secure, and (4) the
interest will be transferable to a purchaser.
The security of title depends largely on the completeness
and reliability of title or ownership data and its evaluation. In
an ideal system, land title records would contain all facts rele-
vant to title, ownership and use.49 In such a system, the
database would be complete and accurate. Consequently, de-
terminations about ownership based on the information in the
database would be reliable, eliminating the need for outside
research.
The attainment of the second principle, speed, depends on
the rapidity and efficiency of data management and evaluation.
For this second goal to be met, data management must include
input, processing, storage and retrieval. Data input should be
swift and administratively simple. Data relevant to a question
of title, ownership or use should be quickly retrievable. Data
evaluation should be as prompt and efficient as possible. Un-
necessary or avoidable evaluations should not be made. Re-
consideration of previous determinations should not be
necessary except in rare circumstances. Irrelevant data would
not be in the database and, therefore, would not be inef-
ficiently appraised.
Realization of the third principle, economy, also depends in
part on the efficiency of data management and evaluation.
Economy results when user costs are reasonably related to the
cost of maintaining and operating the system.5 Where the
government conducts the necessary work, an interest in econ-
omy suggests that users pay a fair share of the cost. However,
the land title assurance system should not be a source of excess
revenue for government. 5 ' The least costly personnel avail-
able who are capable of maintaining the quality of work
needed should perform it.
Most systems have trade-offs. Measures promoting security
of titles may impair economy or efficiency and vice versa. The
object is to design a system that will achieve these three goals
49. See Whitman, supra note 15, at 41-42.

50. Where part or all of the system is operated by private enterprise, this cost
may include a reasonable return for investors.
51. On the other hand, governmental components of such a system should be
self-sustaining Without a subsidy.

as fully as possible with as little conflict between them as is

feasible.
2. Methods of Title Assurance

The two most commonly used methods of land title assur-
ance in the United States are attorney title opinions and title
insurance. With the exception of the few places in the United
States with Torrens registered titles, both methods are based
on recording systems. An attorney title opinion may be based
on a search of the records by the attorney, but opinions are
more commonly based on abstracts of title.5 2 The attorney's
opinion may be expressed in a formal opinion letter.
As title assurance, the attorney title opinion is inadequate.
The opinion is limited to the public record as summarized in
the abstract. Title risks which do not necessarily appear on the
public record, so-called "off-the-record" risks, are not taken
into account, 53 nor is the examining attorney liable if the client
subsequently suffers a loss because of such a risk. 5 4 If the cli-
ent does suffer a loss due to an erroneous title opinion, the
client must prove that the attorney or abstractor was negligent
in order to recover damages. 5 5 Proving that an attorney was
negligent in rendering a title opinion can be difficult.5 6
Inefficiency is another weakness of attorney title opinions.
Attorney opinions often involve repeated evaluations of the
same data each time the title is examined. 5 7 To a limited ex-
52. Abstractors prepare summaries of the public real estate records pertaining to
a particular parcel of real estate. These summaries, called abstracts of title, describe
the recorded prior conveyances or other transactions concerning the subject proper-
ties. The abstract is then presented to an attorney for examination and opinion.
53. See supra note 26 and accompanying text.
54. Off-the-record risks include the very serious possibility of a total failure of
title because the title is based on a void title transaction. See supra note 26 and accom-
panying text for a discussion of the off-the-record risks that may be encountered
under the recording system.
55. HUD/VA REPORT. supra note 31, at III-C-35.
56. In the typical title opinion letter the attorney states whether the title is "mar-
ketable." Marketability is a vague standard and competent title examiners may differ
about whether particular facts appearing in the record negatively affect marketability.
HUD, LEGAL CONSTRAINTS AND REFORMS, supra note 26, at 111-1. An attorney may
opine that a title is marketable. The title may turn out to be unmarketable. If the
attorney was not negligent in rendering the opinion, the attorney will not be liable.
57. This weakness is more pronounced in urban areas where real estate is trans-
ferred more frequently, resulting in longer abstracts summarizing many more trans-
actions than in rural areas. Abstractors typically do not prepare a new abstract from
scratch each time the title is searched and examined. The usual practice is to update
1992]
and Recording:
AND
Land Title Assurance in the Computer Age
RECORDING
tent, this inefficiency is reduced in some localities by reliance

on prior examinations made by other attorneys. Where this
occurs, the examining attorney will generally limit the data
evaluated to that which was added to the abstract after the date
of the previous examination.
The Marketable Title Acts, enacted in eighteen states,5 8 at-
tempt to improve the attorney and abstract system by reducing
the past transactions which must be examined to those re-
corded within a limited period of time, such as the immediately
preceding thirty or forty years, plus an additional time back to
the "root of title."5 9 Claims not preserved by filing within the
limited time period are extinguished by the Marketable Title
Act. While these acts undoubtedly have a positive effect on
an existing abstract. DEP'T OF HUD, LAND TITLE RECORDATION SYSTEMS: ELIMINAT-

ING REPETITIVE TITLE SEARCHES 5 (1979).
Another problem with the attorney opinion method, which is more serious in
areas where real estate is frequently transferred, is that the probability of the title
being found unmarketable or otherwise defective increases as the number of transac-
tions summarized on the abstract grows. Each time real estate is transferred, defects
may arise out of errors or deficiencies in the instrument of transfer, its execution or
recordation. As the number of persons or other entities in the chain of ownership
increases, the possibility increases that a claim, lien, encumbrance or other defect will
attach through them. This problem is exacerbated because the identical data may be
reevaluated each time the title is examined. Examining attorneys are aware that they
do not work in isolation. While an examining attorney personally may believe that
certain data do not amount to a title defect, the attorney may know that other attor-
neys of a more meticulous, conservative bent may conclude otherwise. The examin-
ing attorney does not want to certify title to be good or marketable notwithstanding
certain facts when a future examiner may refuse to so certify because of precisely the
same facts. "Thus the entire process tends to become dominated by overabundant
caution and ultra-meticulous judgments." Paul E. Basye, Trends and Progress-The
Marketable Title Acts, 47 IOWA L. REV. 261, 265 (1962). The problem of overly meticu-
lous title examination caused by the title examiner's awareness that his or her work
may be reviewed by a very conservative, overly meticulous future examiner has been
addressed by the promulgation of title examination standards by bar associations and
others. HUD, LEGAL CONSTRAINTS AND REFORMS, supra note 26, at 13, 111-5; see also
LEWIS M. SIMES & CLARENCE B. TAYLOR, MODEL TITLE STANDARDS (1960). The devel-
opment of standards has helped alleviate the problem:
[Ilt was felt that if certain standards could be laid down in advance, they
could accomplish much to dispel fears that opinions of future examiners
would be at variance with present appraisals. Knowledge as to how others
will treat certain recurring problems will increase the confidence with which
present opinions can be rendered. Thus far real estate title standards have
been adopted in twenty-two states on a statewide basis and in several other
communities on a county or city level. Lawyers in these states unhesitatingly
attest their value.
Basye, supra, at 265.
58. HUD, LEGAL CONSTRAINTS AND REFORMS, supra note 26, at 8, 11-51, II-52A.
59. PAUL E. BASYE, CLEARING LAND TITLES 172 (2d ed. 1970).

WILLIAM MITCHELL IA W REVIEW [Vol. 18
marketability by extinguishing certain old defects, their efficacy

is limited by the exception of certain interests for political, pol-
icy and constitutional reasons.6
The American title insurance industry developed in re-
sponse to the weaknesses in the assurance provided by the re-
cording and attorney opinion systems. When insurance is
requested, a title insurance company searches for and exam-
ines the available data pertaining to the title. Records are kept
of previous searches and examinations, and if the company has
previously insured the title, the company can simply update
the previous search and examination. The data used are
mainly derived from the company's records, a recording sys-
tem or, in a few places, from a Torrens registration system. 6 '
Once the search and examination is completed, the company 62
issues a commitment to issue a final title insurance policy.
The commitment is used as proof of the state of the prospec-
tive transferor's title and contains a list of the possible title de-
fects uncovered by the company's search and examination. By
reporting these defects in the commitment or any policy issued
thereafter, the company excepts them from the coverage of the
policy. Unlike the attorney and abstractor, the title insurer is
absolutely liable for its search and examination, subject to ex-
ceptions and limitations stated in the policy. The typical title
60. Interests and claims of the United States are excepted because of the
Supremacy Clause of the U.S. Constitution. U.S. CONST. art. VI, cl. 2. State govern-
mental claims are also frequently excepted. Most of the Marketable Title Acts except
railroad and public utility easements and easements apparent by user. Several except
leases where the lessee is in possession. A few except possibilities of reverter, rights
of entry, reversions and remainders. BASYE, supra note 59 at 173. The Illinois Mar-
ketable Title Act was held to be inapplicable where there were two competing chains
of title to the same real estate which were each over 40 years old. Exchange Nat'l
Bank v. Lawndale Nat'l Bank, 243 N.E.2d 193, 195-96 (Ill. 1968).
61. The source of the data depends, in part, on the judgment of the company
concerning the economics of data storage and retrieval. The company may rely on
public records in whole or in part. In some urban areas, title companies have organ-
ized elaborate title plants which duplicate public records. HUD, STATE-OF-THE-ART
STUDY, supra note 18, at 41-43. The title plant maintained by the Chicago Title In-
surance Company for Cook County, Illinois, is probably unique in that it not only
completely duplicates the public records, it contains more data than the public
records. The predecessors of Chicago Title Insurance Company were maintaining a
partial private title plant in 1871 which was saved from the Chicago fire which de-
stroyed the County Recorder's records. SHICK & PLOTKIN, supra note 37, at 123-24.
62. In some parts of the United States, this commitment is referred to as a "pre-
liminary title report" or a "binder." See Anthony B. Kuklin, Title Insurance Isn't Every-
thing, in TITLE INSURANCE AND YOU: WHAT EVERY LAWYER SHOULD KNOW! 49, 51
(1979).
19921 McCormack: TorrensTORRENS
and Recording:
RECORDING
insurance policy covers the insured even for losses caused by

certain off-the-record risks6" not covered by the attorney opin-
ion system. The requirements of minimum capitalization and
reinsurance of large risks required by state statutes or regula-
tions assures the property owner that the company will have
funds available to pay a claim. 64
Title insurance is becoming the dominant means of land title
assurance in the United States, 6 5 in part because it is perceived
to be superior to the attorney opinion method. The growing
dominance of title insurance is also due to economies of scale
and promotional efforts of title insurance companies. Further-
more, many national lenders and investors in the secondary
mortgage market insist on relatively standardized title insur-
ance policy coverage 6 6 when they invest in or purchase
mortgages.6 7
The actual difficulties caused by the weaknesses in the re-
cording system as a source of title assurance should not be ex-
aggerated. Losses caused by title defects are fairly
infrequent.6 8 While the recording system may achieve at least
adequate security of title, critics claim that it is too costly and
63. Where a title insurance policy insures the owner of a registered title or an
interest therein, it will typically cover so-called "off-certificate" risks which may be
encountered under Torrens.
64. See S. Gilmer Towell, Insuring the Multi-Million Dollar Transaction-A Major
Lender's Evaluation of a Title Insurer's Balance Sheet, in TITLE INSURANCE AND YOU, supra
note 62, at 245, 249-50; Louis D. Pierce, Reinsurance vs. Coinsurance: What the Title
Customer Should Know, THE GUARANTOR, Sept.-Oct. 1990 (Law. Supp.).
65. See ALLEN AXELROD ET AL., LAND TRANSFER AND FINANCE: CASES AND MATERI-
ALS 764-65 (3d ed. 1986); see also HUD, STATE-OF-THE-ART STUDY, supra note 18, at
38.
66. Frequently, title insurance companies issue policies on standard forms
promulgated by the American Land Title Association (ALTA), a national private as-
sociation of title companies and lawyers. Because of the influence of ALTA, substan-
tial uniformity exists among title insurance policies issued by different title insurance
companies.
67. See AXELROD, supra note 65, at 764-65; see also FEDERAL NAT'L MORTGAGE
ASS'N, SELLING GUIDE, LENDING REQUIREMENTS, CONVENTIONAL FIRST MORTGAGES
214.04 (Oct. 31, 1987) [hereinafter FannieMae] (explaining that the maximum sin-
gle risk assumed by a title company in connection with any mortgage is 50% of the
sum of the company's capital, surplus and reserves).
68. The claims experience of American title insurance companies provides sup-
port. The ratio of claims paid to premiums is very low. Title insurance companies
strive to avoid risks by searching for and reporting defects. With some exceptions,
risks not reported to the insured are covered by the usual title policy. If the record-
ing system were as inadequate as some critics claim, title insurance companies would
suffer far greater claim payments than they actually do.

WILLIAM MITCHELL LAW REVIEW (Vol. 18
inefficient. As a result, some critics have proposed replace-

ment of the recording system with the Torrens system of land
title registration. Yet, these critics have failed to adequately
appreciate that Torrens does not provide a completely reliable
mirror of the true state of title either and that it is more diffi-
cult and costly for government to administer.
III. THE TORRENS TITLE REGISTRATION SYSTEM
A. The Torrens System in Operation: Mirror, Curtain and

Indemnity
The Torrens system of title registration has three major
components: 1) the register of titles, 2) the document vault,
and 3) the assurance or indemnity fund. The basic goal is to
make the governmentally maintained record a conclusive state-
ment of ownership and the condition of title. This conclusive
statement is intended to function as a "mirror" of the true
state of the title and as a "curtain" between the present and
the past which should make it unnecessary to conduct the kind
of historical searches performed in recording systems.6 9
Under Torrens, original certificates of title are maintained
by the administrator of the system, normally called the Regis-
trar of Titles. The certificate names the owner, describes the
property and the estate owned and contains a list of the liens
or encumbrances on the property. 70 Where these liens or en-
cumbrances are created by registered instruments, the original
or a copy of the instrument can be retrieved from the docu-
ment vault or other storage facility for examination.
Upon initial registration or a transfer of ownership, the
transferee receives a duplicate certificate of title.7 ' The
owner's duplicate normally is not kept up-to-date as subse-
quent claims, liens or other encumbrances are registered on
69. Historical searches are sometimes necessary in Torrens. See, e.g., Ivan L.
Head, Real Property--TorrensSystem of Land Registration in Saskatchewan-Exception to In-
defeasibility--Errorin Not Reserving Minerals to Crown Pursuantto OriginalGrant, 34 CANA-
DIAN B. REV. 736, 737-38 (1956). Likewise, historical searches are a necessary part of
the Ontario title registration system. DJ. DONAHUE & P.D. QUINN, REAL ESTATE
PRACTICE IN ONTARIO 20 (4th ed. 1990).
70. See 2 RUFFORD G. PATTON & CARROLL G. PATTON, PATrON ON LAND TITLES
683, 684 (2d ed. 1957).
71. In Cook County, Illinois, the owner must give the Registrar a receipt (called a
"signature card") for the duplicate bearing the owner's signature. ILL. ANN. STAT. ch.
30, 81 (Smith-Hurd Supp. 1991).
the original. The owner's duplicate, or an acceptable substi-

tute, and the deed of conveyance must be presented to register
a transfer of ownership. When an interest is transferred, the
former certificate showing the transferor as owner is canceled,
and a new certificate is issued showing the transferee as the
new owner. Active liens or encumbrances on the old certificate
are carried forward to the new one.
At the time an instrument is presented for registration, the
Torrens office may compare the signature on the instrument
with a specimen of the owner's signature on file. 72 When this
procedure is followed, Torrens has a built-in protection
against the registration of forged instruments. Recording sys-
tems have no comparable protection against recordation of
forgeries.
The assurance or indemnity fund is the third important com-
ponent of the system. 73 Upon initial registration or at some
72. The Cook County Torrens office routinely checks signatures for forgeries;
the Hennepin County, Minnesota, office does not check for forgeries. Interview with
Richard W. Edblom, Examiner of Titles, andJan Witkowski, Registered Property An-
alyst, Office of the Registrar of Titles, Minneapolis, Hennepin County, Minnesota
(Nov. 26, 1990) [hereinafter Edblom-Witkowski Interview]. Arguably, checking sig-
natures is unnecessary when the party initiating registration must present an owner's
duplicate certificate.
73. The coverage afforded by the assurance fund is sometimes likened to title
insurance. However, the two kinds of protection have important differences. First,
title insurance may provide indemnification for insureds when the state of title turns
out to be other than as insured. In contrast, the Torrens assurance fund may provide
compensation for those who lose or fail to attain interests because of errors or mis-
deeds by the registrar or because they are barred by the Torrens act from recovering
interests. See ILL. ANN. STAT. ch. 30, 136-139 (Smith-Hurd Supp. 1991); MASS.
GEN. LAws ANN. ch. 185, 101 (West 1991). The typical title insurance policy will
cover the insured against losses due to off-the-record risks under recording or off-
certificate risks under Torrens. See Michael J. Rooney, A P-imerfor Attorneys, in TrrLE
INSURANCE AND You supra note 62, at 3, 4. The Torrens assurance fund will generally
not provide indemnification for losses caused by off-certificate risks unless registrar
misconduct is involved.
Second, the title insurance and Torrens assurance fund differ in the coverage of
litigation expenses. The typical title insurance policy contains an agreement to de-
fend the title as insured or to indemnify the insured for litigation expenses incurred
because of covered title defects. See id. at 9. In contrast, the Torrens registered
owner or claimant may have to bear alone the cost of title litigation. In Cook County,
pursuant to established policy, the Registrar intervenes in litigation concerning Tor-
rens titles. See, e.g., Echols v. Olsen, 347 N.E.2d 720 (I1. 1976). This policy tends to
supply the title defense coverage available from title insurance.
Third, a Torrens claimant may have to exhaust remedies against others who
might be responsible for the loss before being entitled to claim from the fund. A
lawsuit against the registrar may be required before a claim is paid under some Tor-
rens acts. See, e.g., MAss. GEN. LAWS ANN. ch. 185, 101 & 102 (West 1991), MINN.

time thereafter, the owner or other party may be required to

make a contribution to this fund. The fund is available to com-
pensate those who suffer losses because of errors of the Tor-
rens office personnel and to pay those who wrongfully lose
interests in real estate because of the operation of the
system. 4
Confidence in the Torrens system in California was shaken
in 1937 when the entire state assurance fund was wiped out by
STAT. 508.76 (1990). In other Torrens systems, the registrar may pay claims with-
out being sued. In Cook County, the Registrar may recommend to the county that a
claim should be paid without a lawsuit being necessary. See, e.g., ILL. ANN. STAT. ch.
30, 138-139.1 (Smith-Hurd Supp. 1991). In contrast, a title insurance company
may pay claims without being sued. The insured does not have to exhaust remedies
against others to recover. Where others may be responsible, the insured is simply
obligated to assist the title insurance company in being subrogated to the rights of
the insured against other responsible parties. Of course, title companies may unjusti-
fiably refuse to pay claims.
Fourth, contributory negligence may bar recovery from a Torrens fund. See, e.g.,
MASS. GEN. LAWS ANN. ch. 185, 101 (West 1991). In Illinois, contributory negli-
gence does not bar recovery from the assurance fund where the Registrar commits
gross errors contributing to the loss. See, e.g., Hoffman v. Schroeder, 186 N.E.2d
381, 388 (Ill. App. Ct. 1962) (holding set forth in supplemental opinion). Such negli-
gence does not bar recovery under the terms of the typical title insurance policy.
However, failure to disclose known adverse material facts to the company at the time
of the issuance of the policy may bar recovery.
Fifth, individual claims against a Torrens fund are not limited to a maximum
money amount; title insurance claims are subject to a limit set at the date the policy is
issued (which may be adjusted upward with inflation, in some instances, by optional
inflation endorsements). See MICHAELJ. ROONEY, ATrORNEYS' GUIDE TO TITLE INSUR-
ANCE 14 (1984). In some jurisdictions, insureds under title policies may be able to
recover in excess of policy limits where tortious conduct is proved. See, e.g., Heyd v.
Chicago Title Ins. Co., 354 N.W.2d 154 (Neb. 1984).
Finally, various coverages are available from title insurance companies which
simply are not provided by the Torrens assurance fund. The typical title policy is-
sued currently covers off-certificate risks, such as some statutory overriding interests,
not covered by the Torrens assurance fund and off-the-record risks, such as a void
link in the chain of title, not included in the protection given good faith purchasers
under the recording system. This coverage may include certain tax liens in existence
at the date of policy, right of access to the property, hidden mechanics liens, zoning
classification or violations and encroachments.
In summary, the coverage provided by typical title insurance policies is more
extensive than that available from existing Torrens assurance funds with the impor-
tant exception of the policy limit on claims in title insurance policies. Of course, the
title registration assurance fuitd component could be expanded to include the
broader coverages found in title insurance policies if it is decided that such coverage
should be furnished by government. See generally D. BARLOW BURKE, JR., LAw OF TI-
TLE INSURANCE (1986 & Supp. 1991).
74. See MINN. STAT. 508.76 (1990); ILL. ANN. STAT. ch. 30, 138 (Smith-Hurd
Supp. 1991); MASS. GEN. LAws ANN. ch. 185, $ 101 (West 1991).
1992]
and Recording:
RECORDING
a single claim. 75 Repeal of the California Torrens Act followed

in 1955.76 In Illinois and Massachusetts, local funds are
backed up by the assets and credit of the state or local govern-
ment. 7 In Hawaii and Minnesota, special state funds supply
the back-up.78
B. Initial Registration
Initial registration of title is not a simple matter. The proce-
dure has often been compared to an action to quiet title.79
Once an initial registration becomes final, it has about the
same effect as a final judgment in a quiet title action. Aside
from the important exceptions of off-certificate risks and over-
riding interests, the registration is binding against the whole
world. Indeed, it is often said that the main difference between
recording and Torrens is that in Torrens, the title itself is reg-
istered, while under recording, the evidences of title are
recorded.80
Since initial registration usually involves a binding determi-
nation of the rights in the title being registered, possible claim-
ants must be given due process under state and federal
constitutions. Reasonable efforts must be made to identify and
notify these claimants of the initial registration and they must
be given sufficient opportunity to be heard on their claims.
This process requires title searches and examinations and the
possible expense of preparing and conducting hearings on
claims. Until fairly recently, initial registration under Ameri-
can statutes was handled almost exclusively through judicial
proceedings.81
Initial registration of a title can be expensive. In addition to
75. See Gill v. Johnson, 69 P.2d 1016 (Cal. Dist. Ct. App. 1937).
76. Land Title Law, CAL. GOV'T CODE, foil. 27297.5 (West 1988), repealed by
1955 CAL. STAT. ch. 332, 1.
77. ILL. ANN. STAT. ch. 30, 139.1 (Smith-Hurd Supp. 1991); MAss. GEN. LAWS
ANN. ch. 185, 104 (West 1991).
78. See HAw. REV. STAT. 501-214 (1988); MINN. STAT. 508.75 (1990); see also
Comment, Possessory Title Registration: An Improvement of the Torrens System, 11 Wm.
MITCHELL L. REV. 825, 839 (1985).
79. See, e.g., CRIBBET &JOHNSON, supra note 10, at 353.
80. See, e.g., People v. Mortenson, 88 N.E.2d 35, 38 (Ill. 1949).
81. The first American Torrens act, enacted in Illinois, was held unconstitutional
because it provided for administrative initial registration. A second Illinois act was
passed in 1897 providing for in rem judicial initial registration. See People ex rel.
Deneen v. Simon, 52 N.E. 910, 911 (Ill. 1898).

LAW REVIEW (Vol. 18
WILLIAM MITCHELL
the required contribution to the assurance fund, the title must

be examined to identify possible encumbrances or claimants,
the property may be surveyed, court documents must be filed,
notices must be published, and hearings may be held. Attor-
neys and other experts may be required.
It is difficult to determine precisely the average cost of initial
registration even in a particular location. Attorney fees can
vary considerably. Depending on local practices and proce-
dures, some of the costs may be absorbed by the public, but
most are usually borne by the owner seeking registration.
Where a registration is contested, which rarely occurs, costs
will rapidly escalate. In 1985, Richard W. Edblom, the Exam-
iner of Titles for Hennepin County, Minnesota, estimated the
cost to the registrant of initial registration of a parcel in the
Minneapolis-St. Paul area at between one and two thousand
dollars.8 2 This is roughly in line with published cost figures for
other localities adjusted for inflation.8 3
Regardless of the precise costs of initial registration, the
costs which must be borne by the registrant are likely to sub-
stantially exceed any subsequent savings in seller title transfer
costs due to registration. Therefore, the cost of initial registra-
tion gives the property owner a disincentive to register. This
disincentive is exacerbated where it is likely that title insurance
will have to be purchased in a subsequent sale transaction.
Advocates of registration argue that the cost of initial regis-
tration is justified by the allegedly lower aggregate title and
transfer costs incurred by transferees in subsequent transfers.
While the claim that these aggregate future title costs will be
lower is disputable, it is clear that most of any savings will ac-
crue to subsequent owners and not to the present owner who
must bear the costs of initial registration.
It is generally agreed that the high cost of initial registration
was a primary cause of the failure of Torrens in the United
States.8 4 As a result, three remedies have been proposed: first,
to make registration compulsory; 5 second, to completely or
substantially subsidize the cost of initial registration; or third,
82. Comment, Possessory Title Registration, supra note 78, at 838.

83. In the late 1970s, uncontested initial registrations in the United States cost
about $550 to $750. HUD, LAND TITLE REGISTRATION, supra note 40, at 111-14.
84. See Goldner, supra note 4, at 688.
85. See id. at 695-96 and authorities cited therein; HUD, LAND TITLE REGISTRA-
TION, supra note 40, at V-I, V-8.
to reduce its cost by authorizing an administrative registration

of possessory titles which would become absolute after the pas-
sage of a specific period of time.86
The compulsory approach, given enough time and re-
sources, will lead to increasingly greater use of registration. 7
The important question is whether more widespread use of re-
gistration through compulsion will be justified by the benefits
to society or individuals weighed against the costs to them.
The question of whether compulsory registration is politically
feasible remains unanswered.
Whether the subsidy approach alone will result in increased
privately initiated registration is uncertain. Although initial re-
gistration was substantially subsidized in Cook County, Illi-
nois, during the early years of its Torrens system, there were
only a few hundred registrations per year. These few registra-
tions resulted, at least in part, from the need to establish title
when the Chicago fire of 1871 destroyed the official land title
records. A Chicago ordinance which made registration
mandatory for real estate when acquired by the city was an-
other cause of some initial registrations.8 8
Even if initial registration is free to property owners, de-
pending on the state of local administration, they may be disin-
clined to register if they believe that dealings in Torrens
property are slower or more cumbersome than transactions in
recorded property.89 The existence and use of deregistration
procedures in American Torrens acts indicate that some prop-
erty owners have been willing to bear the cost of removing
their titles from the allegedly superior registration system. 90
86. See HUD, LAND TITLE REGISTRATION, supra note 40, at V-19; Diana Sclar, Min-
nesota Simplifies Land Registration, 11 REAL EST. L.J. 258 (1983); Comment, Possessory
Title Registration, supra note 78, at 840-52.
87. Registration was not successful in England and Wales until it was made com-
pulsory. After over sixty years of effort, many titles in England and Wales are now
registered. See Bostick, supra note 31, at 91 (estimating that over 80% of the land in
England and Wales was registered by 1987).
88. See POWELL, supra note 31, at 131, 145 & 156-57.
89. In 1989, it took over two years before a new certificate was issued when prop-
erty was transferred. See Cook County Blue Ribbon repoort, supra note 10, at 2.
90. In 1983, Illinois added a deregistration provision to its Torrens Act. 1983
Ill. Laws 391. From November 2, 1983, through August 6, 1987, the Registrar of
Titles completed 133 applications for deregistration in spite of the fact that appli-
cants bear most of the costs ofderegistration. FEINSTEIN, supra note 29, at 1-38. Dur-
ing this time, no petitions for initial registration were received. As of June 1989,
there had been no initial registration since 1981. See Cook County Blue Ribbon Re-

Presumably, these property owners perceived some advantage

of recording over Torrens. 9 '
Finally, there is a policy question whether taxpayers in gen-
eral should subsidize an activity which may benefit primarily
only those who frequently deal in real estate.92
C. Administrative Initial Registration of Possessory Titles

In 1982, Minnesota became the first American state to au- 93
thorize administrative initial registration of possessory titles.
These titles become absolute after a certain period of time.94
Apparently, administrative initial registration of possessory ti-
tles is valid under American due process clauses if adequate
notice is given to possible claimants and if the period during
which a possessory title registration could be challenged is suf-
ficiently long.95 The Minnesota authorizing legislation was
heralded as having "the potential for removing the remaining
port, supra note 10, at Appendix G. In 1987, Illinois made deregistration purely ad-
ministrative rather than judicial. FEINSTEIN, supra note 31, at 1-38 to 1-39.
91. In Cook County, developers may deregister when they plan a subdivision
with many lots or a condominium with many units. They wish to avoid the expense
and delay involved with closing many transactions through Torrens. Torrens trans-
actions are closed in an office which has a limited physical capacity. The Torrens
office cannot create many new certificates for new titles in a timely fashion. Develop-
ers also deregister to avoid the expense and delay involved in review of development
documents by Torrens examiners. See FEINSTEIN, supra note 31, at 1-39.
92. 1 AMERICAN LAND TITLE ASSOCIATION, THE TITLE INDUSTRY: WHITE PAPERS,
ch. 5 (1976), reprinted in The Torrens System 56 TITLE NEWS 8 (1977).
93. Act of Mar. 10, 1982, ch. 396, 1982 Minn. Laws 192 (codified as amended at
MINN. STAT. 508A.01-.85 (1990)). Administrative initial registration of possessory
titles has been used for years in England and Australia. U.S. DEP'T OF HUD, IMPROV-
ING LAND TrrLE REGISTRATION SYSTEMS V-19 (1979).
94. This period is five years in Minnesota. MINN. STAT. 508A.01, subd. 2
(1990). Hawaii authorizes registration of possessory titles, but such titles will not
ripen into absolute ones after the passage of a certain period of time. HAW. REV.
STAT. 501-72 (1988). The Minnesota legislation was initially prepared by Mr. Ed-
blom and his staff at the request of a HUD consultant. Edblom-Witkowski Interview,
supra note 70. At the time, Torrens advocates were hopeful that administrative regis-
tration of possessory titles could lead to effective implementation of Torrens
throughout the United States. The attractiveness of possessory title registration
would be greater if titles could be administratively registered without the expense of
searching for, identifying and notifying possible claimants. Such a "pure" form of
possessory title registration probably would be unconstitutional under due process
clauses unless the registration remained inconclusive for a long time such as 10 or 20
years. See id. It is possible that "pure" possessory registrations could be made imme-
diately conclusive if provision is made to pay just compensation to those who lose
interests. This would be a novel application of the power of eminent domain.
95. HUD, LAND TITLE REGISTRATION, supra note 40, at V-9 to V-12; Goldner,
supra note 4, at 700-03.
disincentive for Torrens registration-the high cost of regis-

96
tering uncontested titles by judicial proceeding.
As is the case with judicial registration proceedings, the Min-
nesota possessory title registration statute requires the appli-
cant to submit an abstract of title with the application for
registration.9 7 This is done so that record claimants can be
identified and notified of the registration, thus satisfying due
process requirements. The cost of preparing this abstract
should be about the same as preparing one for a sale under the
attorney opinion method of title assurance. This cost may be a
financial disincentive for owners to register unless they feel
reasonably sure that registration will make their title costs sig-
98
nificantly lower upon a subsequent sale or loan transaction.
Initially, Minnesota's possessory title registration imposed at
least as much cost on the registration office as did judicial re-
gistration. Under both procedures, an examiner of titles ex-
amined and reported on the application and abstract
submitted by the person seeking initial registration.9 9 Upon
enactment of the Minnesota possessory title registration stat-
ute, Examiners of Titles feared that they would be
overburdened by applications for initial registration.' 0 0 Rich-
ard W. Edblom, Examiner of Titles for Hennepin County,
wrote:
Although a court proceeding is not necessary to register ti-
tle under the possessory title registration procedure, it is
still necessary for an examiner to examine the abstract and
write a report stating whether the applicant does have a
good record title .... Consequently, this imposes a work-
load on the staff of our office not much different from a reg-
96. Sclar, supra note 86, at 261.

97. MINN. STAT. 508A.1I (1990). Among other things, the abstract identifies
record interest holders or claimants who should be notified of the proceeding. Hen-
nepin County requires the abstract to "be certified to all of the land to be registered,
as of a recent date, and must include the deed or other instrument by which the
applicant acquired title." Richard W. Edblom, Registration of Title Without a Court
Proceeding 2 (1990) (unpublished report, on file with author).
98. While not as often as in Cook County, Illinois, title insurance is commonly
used in Torrens transactions in Hennepin County, Minnesota. SHICK & PLOTKIN,
99. MINN. STAT. 508A. 13 (1990). Under a "pure" possessory title registration
procedure, the examination and report would not be required. See discussion supra
note 94.
100. Comment, Possessory Title Registration, supra note 78, at 848-49.

ular registration proceeding ... .'0

In response to these concerns, the Minnesota Legislature
amended the possessory title registration procedure so that its
use could be limited, upon recommendation of the county re-
corder and at the option of the county board, to cases where
2
the applicant owned a tract of partially registered land.'1
On August 1, 1990, approximately eight years after passage,
Hennepin10 3 became the first county to adopt the statute.'
While Ramsey 0 5 and Anoka Counties have also expressed an
interest in adopting possessory title initial registration, neither
had done so as of November 1990.106 Representatives of the
Minnesota title insurance industry opposed the adoption of
possessory title registration. 0 7 However, its failure to be
adopted elsewhere in Minnesota may be due to a perception
that it does not significantly reduce the disincentive to the
property owner to register.
The main thrust of the Minnesota possessory title registra-
tion statute is to take initial registration out of the courts and
into the administrative jurisdiction of the Examiner of Titles.
Even if it is assumed that this alone will result in savings, the
101. Id. at 849 n. 191. Since the publication of this statement, the possessory title
registration statute was amended to provide for an examination fee to be paid by the
applicant. MINN. STAT. 508A. 11, subd. 3 (1990). Hennepin County now contracts
with a private law firm to do the title examination. Mr. Edblom wrote on November
26, 1991:
The "examination fee" not only covers the fee charged by the private law
firm, but also covers the overhead of this office. With the examination being
done by other attorneys, there is little work left to be done by our office.
Therefore, the quoted statement greatly distorts the.., procedure now in
operation.
Letter from Richard W. Edblom, Examiner of Title, Hennepin County, to Kathleen
Milner, Staff Member, William Mitchell Law Review 1 (Nov. 26, 1991) [hereinafter Ed-
blota Letter] (on file with the William Mitchell Law Review). This amendment seems to
be inconsistent with the central goal of the possessory title registration statute, which
is to make initial registration more attractive by lowering the cost to the applicant.
102. 1983 Minn. Laws ch. 92, 23 (codified at MINN. STAT. 508A.01(1) (1990)).
103. Hennepin County includes the Minneapolis area.
104. Edblom, supra note 97, at 1. The Hennepin County Board did not choose to
limit eligibility for possessory registration to cases where applicants owned tracts
which were already partially registered. Id.
105. Ramsey County includes St. Paul and the surrounding area.
106. Comment, Possessory Title Registration, supra note 77, at 844 n.144; Edblom-
Witkowski Interview, supra note 94. In addition to Hennepin, another county in
northern Minnesota may have adopted the statute in 1991. Edblom Letter, supra
note 101, at 1.
107. Comment, Possessory Title Registration, supra note 78, at 843; contra Goldner,
supra note 4, at 692 n.152.
costs of identifying claimants to the title and giving them due

process remain. These costs still have to be borne by property
owners or government. The financial disincentive to register
remains if the property owner is assigned a major portion of
these costs. If government assumes such costs, the issue of
whether society should subsidize initial registration is again
material.
Now that one of the larger and better managed title registra-
tion offices in the United States has implemented possessory
title initial registration, it will be interesting to see whether the
frequency of initial registration will significantly increase in
that jurisdiction. Preliminary evidence suggests that it will
08
1
not.
While American Torrens advocates recognize that the cost
of initial registration is a serious barrier to implementation,
most of them fail to appreciate that the quality and security of
registered titles or interests are not as great as they may appear
to be. Usually, a prospective transferee of a registered interest
still needs professional title examination service. Further-
more, while the need for title insurance or its equivalent is not
as great for registered titles as for recorded ones, registered
titles usually involve title risks which may make such coverage
appropriate, if not necessary. As a result, from a cost-benefit
perspective, the incentives for an individual to register and for
government to mandate registration and accept the burdens of
administering registration are not as great as they may initially
appear. While a registered title may appear more secure to a
transferee than a recorded title, the question is whether this
apparent greater security justifies the related costs and
difficulties.
D. Torrens in Operation: Security of Title

Initially, Torrens appears to have an appealing simplicity
compared to recording. Instead of having to search for and
examine the entire recorded legal history of a parcel, it seems
that the title examiner under Torrens simply has to examine
the original certificate of title which "conclusively establishes
108. As of November 26, 1991, 17 applications to register possessory titles in

Hennepin County had been received. Edblom Letter, supra note 101, at 1. Hennepin
County currently has approximately 170,000 unregistered parcels. HUD, STATE-OF-
THE-ART STUDY, supra note 18, at appendix C-4.

the legal status of the parcel's title, subject to some limited ex-
ceptions."' 0 9 These so-called "limited" exceptions are usually
referred to as "off-certificate risks" or "overriding interests."
They make the Torrens register substantially less conclusive
than it is sometimes perceived to be." l0 Except where a party
loses an interest in land to a good faith purchaser from a party
to fraud, generally, those suffering losses from off-certificate
risks are not entitled to indemnification from the assurance
fund unless the registrar's actions constituted misconduct."'
In no title registration system is the register absolutely con-
clusive." 2 Exceptions to conclusiveness found in Torrens sys-
tems are both statutory and judge-made'.. and fall into eight
categories:'"4
1. Caveats: notices on certificates of possible claims or in-
terests which are not technically registered;" 5
2. Governmental Interests: a) rights under federal law in na-
tions with federal systems of government, b) liens or
equivalent interests which secure payment of taxes, c) gov-
ernmental lands for uses such as streets and highways;
3. Private special interest exceptions: a) mechanics liens, b)
judgment creditor or execution liens;
4. Possessory interests: a) short term leases usually where the
lessee is in possession, b) easements by implication, c) other
private easements, although these are rare;
5. Equity: equitable title, due process, and fairness claims
109. Goldner, supra note 4, at 678-79.

110. "[R]egistration is neither as conclusive nor indefeasible in fact as in theory."
M. FRIEDMAN, CONTRACTS AND CONVEYANCES OF REAL PROPERTY 3.12, at 304 (4th
ed. 1984).
111. See ILL. ANN. STAT. ch. 30, 138 (Smith-Hurd Supp. 1991); MASS. ANN. LAWS
ch. 185, 101 (Law. Co-op. 1990); MINN. STAT. 508.76 (1990).
112. The English title registration statute permits a broad range of overriding in-
terests: legal easements, rights relating to adverse possession and "rights of every
person in actual occupation of the land or in receipt of the rents and profits thereof,
save where enquiry is made of such person and the rights are not disclosed." Addi-
tionally, the administrator has authority to change the record by "rectifying" it. Land
Registration Act, 15 & 16 Geo. 5, ch. 21, 70 (1) (Eng.); Bostick, supra note 32, at 94.
113. The existence of some of these excepted claims or interests can, in some
instances, be discovered through other records in the registration office (and else-
where) and through physical inspection of the property.
114. At least some strong or pervasive influences or policies are involved because
these exceptions are not peculiar to title registration. Some are found in recording
systems and marketable title acts as well. Carol M. Rose, Crystals and Mud in Property
Law, 40 STAN. L. REV. 577, 589 n.70 (1988).
115. Illinois abolished caveating in 1979 because it was little used there. 1979 ILL.
LAWS 415, 1 (effective Jan. 1, 1980).
or interests: a) rights to appeal or contest initial registra-

tions, b) exceptions in certain cases of fraud, c) exceptions
to protect the relatively weak or disadvantaged such as con-
tract vendees or the uninformed, d) equitable title interests;
6. Errorexceptions: Exceptions due to errors in administra-
tion of the system;
7. Encroachments: a) encroachments by structures on the
property onto adjacent property or over a building line, b)
encroachments by structures on adjacent property onto the
subject property or over building lines; and
8. Non-title relatedrestrictions on ownership or use: a) planning
laws, b) zoning ordinances or statutes, c) building or hous-
ing codes, d) environmental laws or regulation, e) lack of
access to the property.
1. Caveats
Some Torrens acts provide for placement of caveats on orig-
inal certificates. A caveat gives notice of a possible outstanding
interest or claim. Caveats are aimed at situations where the
validity of an interest is in litigation or otherwise undeter-
mined.i 1 6 A caveat preserves a claimant's rights against subse-
quent transferees while a court, a registrar or other forum
determines the validity of that claim. The caveat is similar to
the lis pendens. Caveating has been described as a recording
system within Torrens, because the government makes no
statement about the validity of the claim. A caveat merely pre-
serves the claim against subsequent purchasers by giving them
notice of it. Although caveats are represented to be inconclu-
sive, they may have an adverse impact on marketability of title.
Unlike recorded instruments, 17
caveats may be removed from
the record by cancellation.i
A caveat explicitly represents the status of the adverse claim
as undetermined. There is another exception to conclusive-
ness which is not explicit. With the exception of overriding
interests and other off-certificate risks, an interest which does
not appear on the register will not be a valid legal interest.i8
Therefore, a claimant normally must register or memorialize a
claim to make it legally valid, but doing so does not necessarily
116. See MAPP, supra note 5, at 143-60.

117. Id. at 148.
118. There might be valid unregistered equitable interests under a court decision
so providing. See Echols v. Olsen, 347 N.E.2d 720, 723 (Ill. 1976).

make it valid. Registered interests are subject to defeasibility

because of overriding interests, exceptions for fraud, notice
exceptions and other exceptions. 19 A Torrens act cannot real-
istically state without exception that all registered interests are
legally valid.
2. Governmental Interests
a. Federal Right Exceptions
Federal claims may take precedence over the register of ti-
tles.' 2 The United States has consented to record (but not to
register) its tax liens and certain other claims to real estate.' 2 '
By statute, in Minnesota and Illinois, federal tax liens are filed
in the recorder's office and not registered. 22 Tax liens are not
a serious problem for title examiners in these jurisdictions. In
Minnesota, title companies will prepare Registered Property
Abstracts which include a list of outstanding federal tax
liens.' 23 The Cook County, Illinois, Torrens office will search
for and report federal and state tax liens when the property is
transferred or when a memorial is entered on the original cer-
119. MAPP, supra note 5, at 99-141. See also Konantz, Koester, McCrossan, and Title
to Torrens Property, 4 WM. MITCHELL L. REV. 59 (1978).
120. Nations having Torrens with federal systems of government include Austra-
lia, Canada and the United States. U.S. CONST. art. VI; MAPP, supra note 5, at 181-82;
Lang, supra note 35, at 205.
121. I.R.C. 6323 (West 1989). See GEORGE E. OSBORNE, HANDBOOK ON THE LAW
OF MORTGAGES 221A (2d ed. 1970). Ironically, the United States government,
which was the source of renewed interest in Torrens in the 1970s, has also been a
barrier to its implementation. The United States has steadfastly refused to register
its claims to real estate held under Torrens acts. HUD, LAND TITLE REGISTRATION,
supra note 40, at V-22. The history of land title record systems reveals a number of
instances of government refusal to perform acts private parties must do to preserve
real estate interests. The governmental refusal is often based on a desire to avoid the
cost of performing these acts. A federally created and related agency, the Federal
National Mortgage Association, is partially responsible for the necessity of title insur-
ing many mortgages on U.S. land (registered and recorded) because of its usual insis-
tence on title insurance on mortgage paper before it will purchase such paper in the
secondary mortgage market. FannieMae, supra note 67, at 36.
122. ILL. ANN. STAT. ch. 30, 84(3) (Smith-Hurd Supp. 1991); MINN. STAT.
508.25(l)-(2) (1990).
123. HUD, LAND TITLE REGISTRATION, supra note 40, at V-22. Apparently, tax
liens and other statutory charges may have presented a greater problem in New
South Wales, Australia and Alberta, Canada, where no provision was made for re-
cording or registering them in the land records office. Lang, supra note 33, at 208.
In New South Wales, a measure requiring statutory changes and other interests to be
registered has proved ineffective. Id. In Alberta, overriding interests such as tax
liens are not carried in registers. MAPP, supra note 5, at 184-85.
tificate. 124 Private rights arising under federal law, such as

claims to real estate in bankruptcy proceedings, may also take
precedence over the register even though they are not re-
corded or registered.
b. State Governmental Interest Exceptions

Interests in real estate held by the state may be exempt from
registration. For example, liens securing real estate taxes, spe-
cial assessments or other taxes may be valid although unregis-
tered. 2 5 An exception is justified for real estate taxes that
attach automatically at the beginning of regular periods. Re-
quiring registration of these liens on a multitude of original
certificates under a manual Torrens system would unnecessa-
rily-burden the government.' 26 Once an owner defaults on the
payment of these taxes, however, exemption from registration
appears less justifiable because registration would not be as
burdensome on government. Likewise, exemption of special
assessments which do not attach on a periodic basis and which
encumber only a limited number of parcels is difficult to
justify.
3. Private Special Interest Exceptions

In some Torrens systems, orders of attachment, writs of exe-
cution and judgment liens are valid even if they are not regis-
tered or noted ("caveated") on the original certificates for the
debtors' titles.' 27 In the jurisdictions where notices of these
claims are filed alphabetically or otherwise, their existence
28
should be ascertainable without difficulty.
In a few jurisdictions, the hidden mechanic's lien is a serious,
statutory off-certificate risk.129 A mechanic or person furnish-
124. ILL. ANN. STAT. ch. 30, 84 (Smith-Hurd Supp. 1991).

125. See NORBERT HORN ET AL., GERMAN PRIVATE AND COMMERCIAL LAW: AN INTRO-
DUCTION 181 (1982).
126. Professor Mapp states that the burden might not be great where the records
are computerized. MAPP, supra note 5, at 184.
127. Id. at 192-93; Lamont, supra note 5, 3.48, at 83. In Illinois, judgments do
not become legal liens against real estate until they are memorialized on the original
certificates in the register. ILL. ANN. STAT. ch. 30, 122 (Smith-Hurd Supp. 1991).
128. MAPP, supra note 5, at 193.
129. ILL. ANN. STAT. ch. 30, 84.1 (4) (Smith-Hurd Supp. 1991); MINN. STAT.
508.25(7) (1990); see Fred I. Feinstein & GuerinoJ. Turano, New Legislation Changes
Torrens Act, 68 ILL. B.J. 72, 73 (1979).

ing labor or material to premises has a fixed period of time 3 '

after the completion of the work to perfect a mechanic's lien
and secure payment by filing in the recorder's or Torrens of-
fice. The priority of a lien that has been properly filed or regis-
tered will relate back to an earlier date.13 ' As a result, a real
estate purchaser in these jurisdictions, under either recording
or Torrens systems, runs the risk that an outstanding unre-
corded or unregistered mechanic's lien may exist which could
be perfected by filing or registration within the prescribed pe-
riod. Physical inspection of the property and affidavits from
sellers, owners or other parties provides some protection
against this risk, but the most effective protection has been title
3 2
insurance.
4. Possessory Interest Exceptions

Short-term leases are frequently exempt from registration
where the lessee is in possession or where there is actual occu-
pation of the property for a period of time set by a state stat-
ute. 33 The justification for this exception is to relieve lessees
130. The period for filing is 120 days in Minnesota. MINN. STAT. 514.08 (1990).
131. The priority will relate back to the date of the commencement of the work in
Minnesota. MINN. STAT. 514.05(1) (1990); Armstrong v. Lally, 209 Minn. 373, 376-
77, 296 N.W. 405, 406 (1941).
132. HUD, LAND TITLE REGISTRATION, supra note 40, at V-25. This risk is serious.
Title insurance companies' aggregate annual claim payments due to losses caused by
mechanics' liens have ranged over a number of years from 8.7% to 22% of total
annual claims paid. Jack Tickner, Mechanics Liens, TITLE NEWS, Jan. 1976, at 40. The
amounts involved in a single claim may easily be many thousands of dollars. The
Uniform Simplification of Land Transfers Act (hereinafter referred to as USOLTA)
proposes elimination of hidden mechanics' liens. Under USOLTA, a mechanic's lien
(called a "construction lien" in USOLTA) will not relate back if the property is "resi-
dential real estate." USOLTA 5-105(b), 5-209; 14 U.L.A. 315, 338 (1990).
'Residential real estate' means ... real estate.. . containing not more than
[3] acres, not more than 4 dwelling units.... A condominium unit that is
otherwise 'residential real estate' remains so even though the common ele-
ments of the condominium . . . include more than [3] acres or the condo-
minium . . . contains more than 4 dwelling units or units used for non-
residential purposes.
14 U.L.A. 315. Purchasers of other kinds of real estate are presumed to be more
sophisticated and, therefore, sufficiently protected by the USOLTA provision which
requires mechanics to file a notice of commencement of work to obtain mechanics'
(construction) liens.
133. The period in Illinois cannot exceed five years. ILL. ANN. STAT. ch. 30, 93
(Smith-Hurd Supp. 1991). In Massachusetts, exempt leases cannot exceed seven
years. MAss. ANN. LAWS ch. 185, 46 (Law. Co-op. 1991). In Minnesota and Al-
berta, leases cannot exceed three years to be exempt. MINN. STAT. 508.25(3)
(1991); MAPP, supra note 5, at 184-85 (citing 64(l)(b) of the Alberta Act). The
19921 McCormack: TorrensTORRENS
and Recording:
RECORDING
and registrars from the administrative burdens involved with

registering leasehold interests that are generally short-term
and of little value. As a practical matter, possessory exceptions
make imperative a physical inspection of property prior to
purchase. Even a short-term lease can have a severe impact on
a purchaser's plans for the future development, sale or use of
property. In Illinois, a physical inspection is also required be-
cause the Illinois Supreme Court has held that an unregistered
but active implied easement of right of way is valid against sub-
land. 34
sequent purchasers of registered
5. Due Process, Notice and Equitable Exceptions

Persons who should have been joined as parties to a registra-
tion but were omitted may challenge the proceedings long af-
ter their completion. 135 Also, Torrens acts generally make the
register inconclusive where a transferee is guilty of fraud or
takes from a party to fraud without consideration paid in good
faith.' 3 6 Courts in some Torrens jurisdictions have introduced
equitable concepts of notice into Sir Robert Torrens' scheme
which he may not have intended. An Australian, G.J. Davies,
analyzed issues of fraud and notice in property registration ac-
3 7
tions in western Canada and New Zealand.1
Davies concludes that Sir Robert intended to keep fraud and
notice separate. Fraud or taking from a party to fraud without
consideration paid in good faith was sufficient to void a regis-
tration in appropriate cases. However, Sir Robert did not in-
tend notice of a prior unregistered interest alone to be
sufficient.' 3 8 Contrary to Sir Robert's intent, courts in New
Australian statutes have excepted leases for periods not exceeding one year in South
Australia and the Northern Territory, not exceeding five years in Western Australia,
and in Victoria without any time limitation when the tenant is in possession. Lang,
supra note 35, at 200.
134. Carter v. Michel, 87 N.E.2d 759, 764 (11. 1949).
135. See Sheaffv. Spindler, 171 N.E. 632 (11. 1930).
136. The case law is split on the effect of transfer by a void instrument or from an
imposter rather than from the registered owner. See Hoffman v. Schroeder, 186
N.E.2d 381, 386-87 (11. App. Ct. 1962); Marcia Neave, Indefeasibility of Title in the Ca-
nadian Context, 26 U. ToRo'rro L.J. 173 (1976).
137. G.J. Davies, Equity, Notice and Fraudin the Torrens System, 10 ALBERTA L. REV.
106 (1971).
138. Sir Robert Torrens' views are reflected in the current Illinois statute. See ILL.
ANN. STAT. ch. 30, 86 (Smith-Hurd Supp. 1991). The statute provides that except
in certain cases of fraud, the registered transferee does not have to inquire into the
circumstances of prior transfers and states: "The knowledge that any unregistered

Zealand, British Columbia, Manitoba and Alberta have treated

notice of prior interests, at least when coupled with an intent
to defeat a prior unregistered interest, as sufficient to consti-
tute fraud. The fraud disqualifies the subsequent transferee
from prevailing over the prior interest. American court 39deci-
sions have also imported notice concepts into Torrens.
6. ErrorExceptions
Administrative errors can create further exceptions to con-
clusiveness. For example, where a memorial erroneously de-
scribes the original document, the original document may
control.140 The transferee, therefore, takes the title as de-
scribed in the original document.' 4 ' Consequently, examina-
tion of the original documents or photographic copies, where
authorized, is imperative. If the registrar erroneously issues
more than one certificate for the same real estate interest, a
situation may arise that is similar to the competing chains of
title problem encountered under recording. In this situation,
courts have held that the earliest certificate prevails. 42 Tor-
rens assurance funds are generally available to compensate
those suffering losses due to such errors.
7. Encroachments
Title defects due to encroachments are of two types: 1) a
structure on the subject property encroaches on adjacent
property or violates a building line on the subject property es-
tablished by a restriction for the benefit of adjacent property;
or 2) a structure on adjacent property encroaches on the sub-
ject property or violates a building line established for benefit
trust, lien, claim, demand or interest is in existence shall not of itself be imputed to
such transferee as fraud." Id.
139. See, e.g., 8930 South Harlem, Ltd. v. Moore, 396 N.E.2d 1, 4 (Ill. 1979) (hold-
ing that principle purpose of a registration system is to provide notice of defects or
encumbrances); County Collector v. Olsen, 362 N.E.2d 1335, 1340 (I11. App. Ct.
1977). Australian and Canadian writers rarely mention Torrens decisions by Ameri-
can courts. This lack of attention is well-deserved. Because few American judges
and lawyers have a basic understanding of title registration, the opinions in the
American Torrens cases tend to be poor.
140. See Horgan v. Sargent, 182 Minn. 100, 233 N.W. 866 (1930).
141. Id.
142. Lamont, supra note 5, 3.45, at 82; see also Turta v. Canadian Pacific Ry. and
Imperial Oil Ltd., 12 W.W.R. 97 (Can. Sup. Ct. 1954).
of the subject property. 43 If the encroachment is substantial,

it creates an encumbrance which may render title
44
unmarketable.1
Encroachments are off-the-record risks under recording sys-
tems and off-certificate risks under Torrens. They do not have
to be recorded or registered to be valid defects. A recording
or registration system cannot move structures or make them
disappear. An examination of a current survey of the property
can provide a real estate purchaser the best protection against
encroachments. If the encroachment is against the subject
property, the owner must demand its removal or accept its
presence. The owner of the structure may have a right to
maintain the encroachment under the doctrine of adverse pos-
session. However, if the property encroached upon is in Tor-
45
rens, adverse possession does not apply.1
8. Traditionally Non-Title-Related Exceptions

Restrictions on ownership and use created by planning, zon-
ing, building and environmental laws traditionally have not
been considered title matters, although violations have some-
times been considered title defects. 14 6 Existing recording or
title registration systems do not include data on land use legis-
lation that restricts ownership even though such restrictions
may have a dramatic impact on the value of real estate. By the
1960s, with powerful computer technology widely available,
proposals surfaced to create computer data banks to store and
retrieve data on real estate parcels. 4 7 Some of these proposals
recommended the inclusion of data on applicable planning,
zoning and similar laws not contained in existing recording or
title registration systems. 48 It is questionable whether land ti-
tle records should contain information about planning, zoning
and similar laws. The experience of the title insurance indus-
try is instructive.
143. 2 PATRON & PATrON, supra note 20, at 676.

144. Id.
145. See ILL. ANN. STAT. ch. 30, 85 (Smith-Hurd Supp. 1991). However, there
might be a right to maintain it under an implied easement if such easements are
exempt from registration.
146. See Richard H. Howlett, Report of Title Insurance Standard Forms Committee, TITLE
NEWS, Jan. 1962, at 63, 65.
147. See Robert N. Cook, Land Data Systems: The Next Steps, 43 U. CIN. L. REV. 527,
527-38 (1974).
148. For a fairly recent proposal, see LANG, supra note 35, at 209-13.

William Mitchell
WILLIAM Law Review,
MITCHELLVol. 18,LA
Iss.W1 [1992],
REVIEW Art. 4 [Vol. 18
Standard American Land Title Association policies exclude

from coverage any losses caused by a "law, ordinance or gov-
ernmental regulation (including, but not limited to, building
and zoning ordinances... ).""9 Under pressure from lending
interests, the title insurance industry agreed to offer limited
zoning coverage by endorsement to policies.150 The endorse-
ment insures that the owner can use the property for a specific
purpose and that improvements meet the requirements of the
applicable zoning ordinance.' 5 1 Title insurance companies
were reluctant to provide zoning coverage because of difficul-
ties in determining the zoning classification and the provisions
applicable to specific properties. 152 If title insurance compa-
nies find that zoning coverage involves personnel and adminis-
trative problems, government should anticipate similar
problems and accompanying costs if it stores zoning and other
1 53
land use information for particular parcels.
9. Conclusion: Security of Torrens Titles

Even with the exceptions discussed, the typical register is
probably more reliable and, thus, promotes security of titles
more than the record in the typical recording system. In other
words, the Torrens system does provide a better mirror of the
title than does the recording system. It also provides better,
though no means perfect, protection against non-record inter-
ests. Nonetheless, the belief that registration systems may be
capable of eliminating the roles of title insurance companies or
attorneys in modern conveyancing is probably overly optimis-
tic. The prudent purchaser of a registered interest will still
need professional title services. The exceptions to the conclu-
siveness of the register may make title insurance advisable in
149. Usually, this provision is found in part one of the "Exclusions From Cover-
age" in standard A.L.T.A. policies. D. BARLOW BURKE, JR., LAW OF TITLE INSURANCE
3.2 (1986).
150. See AXELROD ET AL., supra note 65, at 798-99.
151. BURKE, supra note 149, at 418-19 & n.3 (referencing ALTA Endorsement
Form 3.1).
152. See Charles C. Smith, Title Insurance and Zoning Coverage, THE GUARANTOR,
Spring 1976, at 3. These difficulties included retaining persons competent to sort
through complex zoning ordinances and the poor record-keeping of some municipal-
ities. Id. Smith also suggests that providing zoning coverage might also constitute
the unauthorized practice of law. Id.
153. Id.
19921 McCormack: Torrens TORRENS
and Recording:
ANDLandRECORDING
Title Assurance in the Computer Age
many cases. 1 54
E. User Difficulties with Torrens

Initially, Torrens appears to be more user friendly than re-
cording. It is obviously less burdensome for the user to ex-
amine the original certificate of title in the register than it is to
wade through an often complicated recorded chain of title.
While title examiners may find a registration system less pain-
ful and arduous, the parties initiating registration may have a
quite different experience. The rigidity and inflexibility of a
Torrens office can be frustrating and time-consuming.
Under recording, users ordinarily evaluate title data when-
ever they deem it necessary. These evaluations do not abso-
lutely control whether data is added to the record. The
recorder's acceptance of data for recording does not subject
government to potential liability from an erroneous evaluation
because the recorder usually makes no evaluations. In con-
trast, title registration operates as a piecemeal, continuing,
quiet title action, albeit without all of the due process require-
ments applicable to initial registration.' 55 Government makes
the evaluations one at a time, and they usually are conclusive.
The act of acceptance includes a governmental judgment that
the instrument was effective to create, transfer, modify or can-
cel the interest referred to therein. Consequently, the govern-
ment may be liable for erroneous or wrongful evaluations.
Under these circumstances, it is not surprising that registrars
are frequently quite careful about accepting instruments for
registration.
An Australian lawyer's comment about registrar attitudes in
his country, where initial registration is administrative rather
154. The Torrens assurance fund usually does not provide compensation for
losses caused by off-certificate risks. See, e.g., ILL. ANN. STAT. ch. 30, 138 (Smith-
Hurd Supp. 1991); MINN. STAT. 508.76 (1990). However, in Shevlin-Mathieu Lum-
ber Co. v. Fogarty, 130 Minn. 456, 153 N.W. 871 (1915), the court held that the
failure of the Registrar to report the existence of an overriding U.S. government
interest was an "omission," within the meaning of the Minnesota statute, entitling
the insured purchaser to recover from the assurance fund. Id. at 460, 153 N.W. at
873. Another serious risk is lack of access to the property, which can result in cata-
strophic loss and is usually covered by title insurance and not covered by the Torrens
assurance fund. See, e.g., L. Smirlock Realty Corp. v. Title Guarantee Co., 418 N.E.2d
650, 652 (N.Y. 1981) (rendering real estate worth over $600,000 valueless from lack
of access).
155. See, e.g., Eliason v. Wilborn, 281 U.S. 457 (1930).

William Mitchell LawMITCHELL
Review, Vol. 18, Iss. 1 [1992], Art. 4
LAW REVIEW [Vol. 18
WILLIAM
than judicial, is pertinent here even though the lawyer was spe-
cifically referring to acceptance of applications for initial
registration:
One of the principal criticisms of the administration of the
Australian Torrens acts . . . has been that registrars have
been far too finicky in scrutinizing applications for registra-
tion ... 56
These failures on the part of the administrators of

the Torrens scheme do not reveal a fundamental ineffi-
ciency or incompetence on their part, but simply a lack of
imagination in seeking the most effective means of imple-
menting the goals of the scheme. They have not been able
to appreciate that their duty does not lie in protecting the
assurance fund at all costs, but also in broadening the reach
of the scheme, and their real duty lies in achieving
57 a sensi-
ble accommodation between these two aims.'
Critics of Torrens have referred to registrar inflexibility as a
problem, usually without describing or even citing specific
examples. 158
156. Ronald Sackville, Some Aspects of the Torrens System in Victoria: A Com-
parison 37 (1966) (unpublished manuscript on file with the Yale Law School Library
collection), quoted in AXELROD ET AL., supra note 65, at 908-09.
157. Id. Registrar zeal in protecting the indemnity or assurance fund reached a
zenith (or is it a nadir?) in Hoffman v. Schroeder, 186 N.E.2d 381 (Il. Ct. App.
1962). In this case, the court held that a savings and loan which took a forged mort-
gage from an imposter did not acquire a lien against the property even though a
memorial of the mortgage was registered. Id. at 388. The court said that the Illinois
Torrens Act protects those who take from a party to fraud only where they take from
a registered owner in good faith. Id. at 386. Since the savings and loan took from an
imposter and since it lacked good faith because its officer improperly notarized the
mortgage, no lien attached to the subject property. Id. However, the court held that
the savings and loan was entitled to indemnification because of the Registrar's failure
to discover the "palpable" forgeries on the mortgage documents. Id. at 388. In his
petition for rehearing, the Registrar cited an 1889 New Zealand decision holding that
contributory negligence may bar a party from recovering from the assurance fund.
Id. at 388-89 (citing Miller v. Davy, 7 N.Z.L.R. 515 (P.C. 1889). The Registrar argued
that the improper notarization by the officer of the savings and loan was contributory
negligence which barred recovery from the assurance fund. The court rejected this
argument:
As we pointed out in our opinion, the indemnity fund is in the nature of
insurance against mistakes of the registrar .... To import into this type of
transaction the doctrine of contributory negligence, whereby gross errors of
a registrar might be completely cancelled out by some slight mistake on the
part of an insured, could discredit the whole system and destroy it as an
effective means of assuring titles in this county.
Id. at 389.
158. See Risk, supra note 5, at 478.
Instruments are examined when they are presented for re-

gistration. If the examiner believes the instrument is irregular
or defective in some respect, the examiner will refuse registra-
tion.' 59 For example, at some offices, the examiner may refuse
registration if the signature appears to be forged. Registration
of a forged instrument may cause the true owner to lose title to
the land.' 6 ' To satisfy the registrar, a proceeding subsequent
to registration may be necessary or, at a minimum, the trans-
ferring owner may have to appear in the Torrens office to ac-
knowledge the authenticity of the signature and to prove
identity. 16,
Examiners may also refuse to accept instruments which con-
tain minor discrepancies in the description of the parties or the
property. To satisfy the registrar, an owner may have to con-
tinue an error or resort to court or administrative proceedings
159. This examination process, of course, has the advantage of detecting potential
defects before they are entered into the record. A similar method of quality control
exists in the Scottish recording system. There, the recorder (called the "Keeper of
the Public Register") examines the instruments presented for recordation. See Fiflis,
Security and Economy, supra note 2, at 189.
Under this process of examination, mistakes are caught, not buried, as is the case
under the American recording system. This feature of the Torrens system may not
be worth the cost to government and the cost and aggravation to the users of the
system. Adjournment of real estate closings and repeated trips to the government
office until the bureaucrats are satisfied are real costs, although difficult to measure.
Few real defects are caught. The Registrar's objections often appear to be bureau-
cratic nitpicking.
Years ago, scandals occasionally arose in the Torrens office in Cook County,
Illinois, involving the bribing of counterpersons. In some instances, it appears that
illegal payments were used to inject flexibility into the system. See United States v.
Gannon, 684 F.2d 433 (7th Cir.) (en banc) (affirming conviction of counterman at
Cook County Recorder of Deeds office for accepting payments in excess of what law
required), cert. denied, 454 U.S. 940 (1981); William Juneau, Boss Defends Bribed Aide as
'a Bit Guilty', CHI. TRIB., Sept. 25, 1980, at 1, 22 (quoting Cook County Recorder of
Deeds). Presently, a cleaner regime prevails. Since 1977, the following notice has
been displayed conspicuously at several places in the Torrens office: "ATTEN-
TION-Office regulations prohibit members of the staff from accepting gratuities
from any source in the conduct of office business." Gannon, 684 F.2d at 435; see also
FEINSTEIN, supra note 31, at 1-43.
160. See Eliason v. Wilborn, 281 U.S. 457, 459 (1930).
161. The author once represented a person who owned a Torrens registered con-
dominium unit. He wanted to convey the unit to his fiancee and himself in joint
tenancy. The Cook County Registrar refused to accept the deed which was prepared
to accomplish that result because the Torrens office workers suspected that the signa-
ture might have been forged. The client was accordingly required to appear in per-
son with two pieces of identification to prove his identity and acknowledge the
execution of the deed.

to correct it.' 6 2 Such administrative inflexibility inconve-

niences the parties to a Torrens transaction 63 and creates ad-
ditional user costs, such as time expenditures and added
closing costs.
The inflexibility which may be encountered in Torrens sys-
tems causes other problems as well. For example, registrars
may refuse to remove memorials of encumbrances which no
longer exist, even those that cannot reasonably still exist.164
Memorials may render the title unmarketable and may only be
eliminated through a subsequent proceeding in court.16 5 This
process is costly and further undermines the owner's confi-
dence in Torrens. At times, Torrens systems also require sub-
sequent proceedings in court to establish title or the transfer
of title. For example, upon the death of a registered owner,
the Torrens office may require a subsequent proceeding sepa-
rate from any probate proceedings in order to establish devo-
66
lution of ownership.1
162. See Workshop on the Torrens System: A Panel Discussion, TrrILE NEws,Jan. 1977, at
34. For example, where the name or marital status of an owner has changed since
the previous registration, a proceeding subsequent to registration may be necessary
to establish the owner's identity or claim to ownership. Id. By contrast, the Registrar
in Cook County, Illinois, normally will accept affidavits to establish identities or
claims to ownership after legal name changes or changes in marital status. As a re-
sult, subsequent court proceedings are usually unnecessary.
163. Torrens closings in Cook County must take place in the Torrens office be-
cause the paperwork has to be scrutinized by the employees or deputies of the Regis-
trar. Members of the general public must reserve closing space at least a month in
advance. See Cook County Blue Ribbon Report, supra note 10, at 44.
164. See Workshop on the Torrens System: A Panel Discussion, supra note 162, at 34.
165. In an attempt to cure this problem, a 1979 amendment to the Illinois Tor-
rens Act authorized the Registrar to cancel memorials claimed no longer to be valid
in return for a title indemnity agreement backed by a surety bond from a title insur-
ance or bonding company or other security. If the Registrar later paid a claim from
the assurance fund because the cancellation was wrongful, the Registrar would be
entitled to indemnification under the title indemnity agreement. See ILL. ANN. STAT.
ch. 30, 109.1 (Smith-Hurd Supp. 1991). Previously, the established title insurance
practice had been to furnish a title insurance company with a title indemnity bond
(frequently issued by itself or a related company) to induce the company to remove
an exception to coverage based on a title defect of questionable validity.
166. See R.G. Patton, Evolution of Legislation on Proof of Title to Land, 30 WASH. L.
REV. 224, 234 (1955). In Illinois, this kind of proceeding was called a "transmission
proceeding." In the early 1970s, the Registrar in Cook County, without statutory
authority, began transferring certificates without court orders from transmission pro-
ceedings. If the estate had been probated, the Registrar relied on the determinations
made by the probate court. If no proceedings had been conducted in probate court,
the Registrar administratively determined the devolution of the property and accord-
ingly reregistered the property. This chagrined the judges who normally conducted
transmission proceedings. In 1977, the Illinois Legislature amended sections 71 and
Title insurance companies and title examiners under the at-

torney opinion method of title assurance are often inflexible
and overly conservative, but their pronouncements are not
backed up by the state. In addition, title insurance companies
will generally retract their objections after the owner furnishes
statements, affidavits, title indemnity bonds or pays an addi-
tional premium. All of these items are usually less costly and
more expeditious than getting court or administrative or-
ders. 16 7 Finally, an unsatisfied owner usually has the option of
using another company or examiner. An owner cannot choose
another Torrens office.
F. Operation and Administration of Torrens: Economy and

Efficiency ?
At their extreme, registration advocates paint the private ti-
tle industry as populated by parasites feeding off the inefficien-
cies of the recording system. 168 The advocates claim that
replacement of recording with Torrens registration will result
in substantial savings to the public through the elimination or
reduction of title examination or insurance costs. However,
the comparative public and private costs of different land title
assurance systems are difficult to assess. The available data are
sporadic, incomplete and, in some instances, unreliable.
Even if reliable data were readily available, meaningful quan-
tification of the relevant costs presents difficult analytical
problems. Despite the well-financed and in-depth studies of
recordation and title registration in the United States done in
the 1970s by Arthur D. Little, Inc.,' 69 and Booz, Allen & Ham-
ilton, Inc., 170 a statement published in 1966 by Professor Ted
72 of the Illinois Torrens Act to authorize the Registrar's creative way of enabling
owners to avoid transmission proceedings. See ILL. ANN. STAT. ch. 30, 115-16
(Smith-Hurd Supp. 1991).
167. In Cook County, the Registrar has developed a set of affidavits which may
suffice to cure objections without going to court. Additionally, transferee title objec-
tions to Torrens titles can often be satisfied by title insurance coverage.
168. See, e.g., Myres S. McDougal &John W. Brabner-Smith, Land Title Transfer: A
Regression, 48 YALE L.J. 1125, 1142 (1939).
169. The Arthur D. Little study was conducted mainly for the American Land Ti-
tle Association. See SHIcE & PLOTKIN, supra note 37.
170. The reports include the following: LAND TITLE SYSTEMS: LEGAL BIBLIOGRA-
PHY (1978), LAND TITLE RECORDATION SYSTEMS: LEGAL CONSTRAINTS AND REFORMS
(1979), LAND TITLE RECORDATION SYSTEMS: ELIMINATING REPETITIVE TITLE SEARCHES
(1979) and LAND TITLE REGISTRATION SYSTEMS: PROPOSALS FOR IMPROVEMENT (1979)
(also published by the U.S. Department of Commerce National Technical Informa-

Fiflis remains true today: "There is no reliable, current analysis

of the comparative out-of-pocket costs under the title registra-
tion, recording, and title insurance systems."'' Thus, the eco-
nomics of Torrens in relation to recordation are difficult to
quantify and evaluate.
Several published cost comparisons fail to distinguish be-
tween two basically different cost problems. One problem has
to do with the resources expended to operate the systems.
Here the issue is one of comparative economic efficiency. The
other problem is about the fairness of the charges borne by
consumers. Here the focus should be on whether consumers
are paying for unnecessary or avoidable services or whether
they are paying too much for them. For example, legal fees for
title examinations may not be indicative of actual resources ex-
pended in such examinations. If lawyers charge equivalent
amounts for examining Torrens and recorded titles, that does
not necessarily demonstrate that the efficiency and economic
cost of examination is equivalent in these systems. It may sim-
ply mean that lawyers charge fees for examinations unrelated
to the actual time and resources expended. If the premise is
that fees should be related to the time and resources ex-
pended, it may be that some consumers are paying too much
for title examination services.
Professor Risk suggests two relevant costs: costs to govern-
ment and costs to users. 72 Costs to government are relatively
easy to ascertain. 73 User costs present more difficulty since
they include the costs of searching for, examining, and solving
problems uncovered by searches and closing. These costs, ac-
cording to Professor Risk, should not be measured by charges
made by government or service providers to users and cli-
ents.174 Risk appears to believe, as the author does, that these
charges are not systemic but rather external and do not accu-
rately reflect the actual resources expended.
tion Service under the title IMPROVING LAND TITLE REGISTRATION SYSTEMS). Two
additional reports include LAND TITLE RECORDATION PRACTICES: A STATE-OF-THE-ART
STUDY (1980) and LAND TITLE MAPPING AND SURVEYING (1981).
171. Fiflis, Land Transfer Improvement, supra note 7, at 458.
172. Risk, supra note 5, at 477.
173. Id. Such costs include the costs of staff, space and equipment and can thus be
determined by accounting records and investigation. Id.
174. Id. Risk criticizes studies by Professors Powell and Fiflis comparing costs
based in part on charges made to users and clients. Id.at 477 n.54.
Professor Fiflis attempted to make a rough comparison of

user costs under recording and Torrens systems by using at-
torney fee schedules, title insurance rates, recording and regis-
tration fees, and questionnaires sent to title insurance
companies, abstractors and attorneys.1 75 He concluded: "All
in all, it seems that transfers of registered titles are likely to be
much less costly than transfers of recorded titles, especially
when the cost of title insurance is included."'' 76 Professor Fif-
lis' conclusion is flawed by the common erroneous assumption
that title insurance costs are appropriately incurred only with
recorded titles. 7 7 He fails to consider title insurance costs in
Torrens transactions that are title insured. More importantly,
Fiflis' data on comparative governmental costs in operating the
systems is limited
8
to the costs reflected by registration or re-
cording fees. 17
In the mid-1970s, Shick and Plotkin studied Torrens systems
in Illinois, Massachusetts and Minnesota.' 79 They attempted
175. Fiflis, Land Transfer Improvements, supra note 7, at 458-68. Professor Powell
also made cost comparisons in his study for the New York Law Society. Because his
data are now so old and because of flaws in his research design, as pointed out by
Professor McDougal and others, Powell's cost comparisons are not considered here.
See POWELL, supra note 31; see generally McDougal & Brabner-Smith, supra note 168.
176. Fiflis, Land Transfer Improvement, supra note 7, at 468.
177. Professor Fiflis compared the following fees: attorney's, abstractor's and gov-
ernmental recording and title registration. He also attempted to compare convey-
ancing costs in a number of special situations. Fiflis concluded that there was no
difference in attorneys' fees in Minnesota and Illinois and that buyers' attorneys' fees
were one-third to one-half for registered titles in Massachusetts. Surveyors' fees
were the same. Registration fees were slightly higher than recording charges. Ab-
stractors' fees were a significant cost under recording and not incurred under regis-
tration. Fiflis also treated title insurance costs as applicable solely to recording. See
id. at 458-62.
It is possible that insuring Torrens titles was uncommon when Professor Fiflis
collected his data, but that is not true today. While this is due at least in part to
demands from investors in the secondary mortgage market, title insurance provides
protection against potential losses due to overriding interests or off-certificate risks
and other causes not covered by the usual Torrens assurance fund. Fiflis was aware
of these risks or interests: "[M]ost of them can be determined conclusively by further
action, such as examination of additional records, inspection of the premises, and
inquiry of occupiers." Id. at 452. If this is true, Fiflis should have taken into account
the costs of making these additional investigations in Torrens transactions to make a
fairer and more accurate comparison.
178. Professor Risk criticizes Fiflis and Powell because "no distinction is made
between the costs of operation and the costs of change, and because the costs are
assumed to be the costs charged to the users and clients." Risk, supra note 5, at 477
n.54.
179. See SHICK & PLOTKIN, supra note 37, at 56.

to ascertain and compare closing costs under Torrens and re-

cording in those jurisdictions.' In Cook County, Illinois,
where title insurance is almost the sole form of land title assur-
ance, Shick and Plotkin found that Torrens transfers cost less
than recorded transfers if the property was more expensive
housing.' 8 ' However, for less expensive housing, uninsured
Torrens transfers were more costly.' 8 2 Where a Torrens trans-
fer was title-insured, closing costs were substantially higher
83
than those incurred in insured transfers under recording.1
Shick and Plotkin found different results in Massachusetts.
There, a common form of land title assurance was the attorney
opinion method based on attorney searches rather than ab-
stracts. 184 The authors found little or no difference in closing
costs between registered and unregistered parcels. 8 5 In Hen-
nepin and Ramsey Counties in Minnesota, Shick and Plotkin
reported that title owners frequently obtained title insurance
for both registered and unregistered property.' 86 They also
found that registration is not perceived as having any signifi-
8 7
cant effect on closing costs at the time of resale.'
Because Shick and Plotkin received support from the Ameri-
can Land Title Association, some critics have questioned their
objectivity.'8 8 Moreover, their cost figures are based on fees
180. See id. at 142-44.

181. Id. at 143.
182. Id.
183. Id.
184. Id. at 102.
185. Id. at 119.
186. Id. at 97.
187. Id. at 96. In contrast to the findings and conclusions of Shick and Plotkin, the
report on improving land title registration systems prepared for HUD by Lane and
Edson, a Washington, D.C., law firm, asserts "that the cost of a title search and exam-
ination can be substantially reduced, and the cost of title insurance eliminated, under
a registration system." HUD, LAND TITLE REGISTRATION, supra note 40, at 11-2. To
support this assertion, Lane and Edson do not cite empirical evidence. Instead, they
refer to the study byJanczyk, see infra note 190, and an article by Professor Whitman,
see supra note 15. These authorities do not provide adequate support for the asser-
tion. Contrary to Lane and Edson's assertion, off-certificate risks in existing Torrens
systems are serious enough to make title insurance frequently advisable, if not neces-
sary. The typical Torrens assurance fund presently is not a substitute for the cover-
age available under ALTA title insurance policies. However, registration systems
could provide protection equivalent to that currently available from title insurance.
188. See, e.g., CUNNINGHAM, supra note 23, 11.15, at 828 n.5; JOHN E. CRIBBET ET
AL., CASES AND MATERIALS ON PROPERTY 1366 (6th ed. 1990).
and charges which are not reliably indicative of actual compar-

ative economic efficiency.
Two published studies attempt to compare the economic ef-
ficiency of title registration and recording in actual operation.
One was conducted under the direction of the Ontario Law
Reform Commission; 89 the other was by an economist, Profes-
sor Joseph T. Janczyk. 190 Janczyk attempted "to estimate the
social opportunity costs directly from the expenditures of the
institutions involved."' 19 ' His study in Cook County, Illinois,
compared the costs of operating the Torrens system with the
costs of running the recording system. The costs of the re-
cording system were derived from the expenditures of the
Cook County Recorder's Office, Chicago Title and Trust Com-
pany, processors of title transfer documents and surveyors.
The costs of the Torrens system were based on the expendi-
tures of the Cook County Torrens Department and processors
of title transfer documents. 192 His cost data were obtained
from interviews, reports from the Cook County Recorder/Reg-
istrar of Titles, and from Chicago Title and Trust Company.
Janczyk concluded that the title transfer costs in Torrens were
approximately $100 (in 1967 dollars) less than in recording.'
He determined that significant cost savings would be achieved
through a complete conversion of recorded parcels in Cook
County to the Torrens system. 9 4 Janczyk recommended that
"Cook County seriously consider enacting legislation requir-
ing that property currently being transferred in the recording
' 95
system be switched into the Torrens system."'
Janczyk's study is unpersuasive because of the apparent un-
189. ONTARIO LAW REFORM COMMISSION, DEP'T OF JUSTICE, REPORT ON LAND RE-
GISTRATION (1971) [hereinafter ONTARIO LAND REGISTRATION REPORT].
190. Joseph T. Janczyk, An Economic Analysis of the Land Title Systems for Transferring
Real Property, 6J. LEGAL STUD. 213 (1977). Janczyk's study is more readily available in
the United States than the Commission's report. Accordingly, Janczyk's study has
received more attention in this country.
191. Id. at 216.
192. Id. at 217.
193. Id. at 215.
194. Id. at 226.
195. Id. Contrary to Janczyk's suggestion, Recorder/Registrar Carol Moseley
Braun recommended abolition of the Cook County Torrens system and lobbied the
Illinois Legislature to pass the necessary legislation, which was enacted in January
1991. See Cook County Blue Ribbon Report, supra note 10; Recorder Urges End to Tor-
rens System, CHI. TRIB., May 31, 1990, 2, at 6; Mark Hornung, Best-laid Plans Fall Far
Short in Recorder's Office Revamp, CRAIN'S CHI. Bus., Mar. 5, 1990, at 56.

reliability of some of the data he used as well as serious factual

misconceptions about the operation of conveyancing and land
title assurance systems in Cook County. One of the most seri-
ous errors he made was assigning the entire cost of Chicago
Title Insurance Company Cook County title service operations
to the recording system.' 96 A common misconception is that
registration makes title insurance unnecessary and that regis-
tered titles are rarely, if ever, insured.' 97 Chicago Title Insur-
ance Company, for example, has been insuring substantial
numbers of Torrens transactions in Cook County since at least
1937.198 An estimated ninety-five percent of Torrens transfers
in Cook County currently involve title insurance.' 99 Indeed,
deputized employees on the payroll of Chicago Title Insurance
Company work side by side in the Torrens office with Cook
County employees. 2 0 Therefore, it was erroneous to attribute
the entire cost of Chicago Title Insurance Company Cook
County title service operations to its recording system, instead
of assigning some portion of the costs to Torrens.
Janczyk's study also fails to adequately take into account im-
portant systemic user costs. 20 ' These include the costs of ex-
amining the register or title insurance reports, making and
examining surveys, clearing title objections under both Tor-
rens and recording, and closing transactions. While Jancyzk
does consider these costs, he treats them as identical in both
196. Janczyk, supra note 190, at 227. Other errors include his statement that Chi-
cago Title Insurance was the "only title insurance firm in Cook County." Id. As an
apparent afterthought, he states that Pioneer Title Insurance was a "minor excep-
tion." Id. at 227 n.32. Because of Chicago Title's dominance, the U.S. Justice De-
partment sued under the antitrust laws in the 1970s. See Chicago Title and Trust Sued,
CHI. TRIB., Feb. 6, 1973, at 1. Chicago Title settled with the United States. Cur-
rently, approximately 70 title companies do business in Cook County with Chicago
Title Insurance doing about 38% of the business. See Who Cares About Title Insurance?,
CHI. TRIB., Apr. 30, 1989, 16, at 1.
197. See, e.g., Janczyk, supra note 190, at 214.
198. See POWELL, supra note 31, at 143 n.90.
199. Cook County Blue Ribbon Report, supra note 10, at 52. Shick and Plotkin
reported that Chicago Title Insurance Company (which at the time had 70 to 80% of
the title insurance business in Cook County) insured a total of 1,833 of the 12,866
Torrens transfers in 1976 and 2,059 of 9,650 transfers in 1975. They concluded that
this data meant that as much as 25% of Torrens closings included title insurance
costs. SHICK & PLOTKIN, supra note 35, at 143. If these figures are accurate, the use
of title insurance in Torrens transfers in Cook County has grown substantially in the
last 15 years.
200. Cook County Blue Ribbon Report, supra note 10, at 50.
201. See, e.g., Risk, supra note 5, at 477.
1992]
AND RECORDING
systems, with the exception of surveys, which he assumes are

required in recording and not in Torrens.20 2 To assume that
no need exists for surveys in Torrens is erroneous. Surveys
are used for more than just determining the boundaries of
property. Surveys may be advisable under both Torrens and
recording because they provide the best way to discover en-
croachment defects 20 3 and they usually indicate whether there
is sufficient access to the property. Finally, the most serious
flaw in Janczyk's study is that much of the cost estimation is
based on rough estimates by a few, albeit knowledgeable, indi-
20 4
viduals interviewed.
A study conducted over twenty years ago under the direction
of the Ontario Law Reform Commission is probably the best
empirical cost comparison of title registration and recording
currently available.20 5 Ontario, Canada, is one of the few
places in the world where title registration 20 6 and recording
systems operate side by side in the same localities. At the time
of the Commission study, approximately fifteen percent of On-
tario titles were in the title registration system; the remainder
were in deed registration (recording). 20 7 The Commission at-
tempted to determine comparative user and governmental
costs of the two systems where they were operating side by
side. Instead of measuring user costs on the basis of fees or
charges, the Commission used time expended in searching
title.
The Commission concluded that user searching costs were
higher under the registry system (recording) than the title re-
gistration system. Further, the searches in the recording sys-
tem took longer. The amount of time needed for a search
depended on the experience of the searcher; conveyancers and
students required more time than lawyers. A conveyancer con-
ducting a search in Toronto needed two and one-half hours
more to search the registry system than the title registration
202. Janczyk, supra note 190, at 231.

203. Encroachment defects are not covered by Torrens assurance funds.
204. Janczyk, supra note 190, at 230, 232-33.
205. ONTARIO LAND REGISTRATION REPORT, supra note 189.
206. The Ontario title registration system (called "Land Titles") is not a Torrens
system and does not presently use title certificates. The Ontario Land Titles system
is based on an early version of the English title registration system and not directly
on the Australian Torrens model.
207. ONTARIO LAND REGISTRATION REPORT, supra note 189, at 16.

system. Lawyers who conducted searches in northern Ontario

needed only a half hour more for a registry search. For To-
ronto and southern Ontario, the area in which most of the con-
veyancing is done, a registry search took on average about two
hours longer.20 However, the Ontario Commission also
found that the closing costs were higher under title registra-
tion than for the registry system (recording). In addition, the
government's cost to operate the title registration system was
higher than the costs incurred with recording.
On the basis of the report of the Law Reform Commission
and his own experience, Professor Risk concluded that the ag-
gregate operational costs of recording systems are higher than
in registration systems even though user closing costs and gov-
ernmental costs are higher under registration systems. 20 9 He
reached this conclusion because title searches and examination
under the recording system are longer and more repetitive.
To the extent that repetitive search and examination is not
eliminated by reliance in some fashion on prior searches and
examinations where the same title is being re-examined, Pro-
fessor Risk is probably correct that these user costs are higher
under recording. However, these costs may not be very signifi-
cant because the additional time required to search and ex-
amine title in recording systems is generally not great.2 t0 Also,
computerization can substantially reduce the time expended in
searching under recording. In addition, the Ontario study did
not consider the effect of title insurance on comparative costs
21
because Ontario did not use title insurance at the time. 1
While the comparative user costs of Torrens and recording
have not been satisfactorily determined and measured, govern-
mental costs of operation are another matter. In contrast to
208. Id. at 22.

209. Risk, supra note 5, at 477-78.
210. ONTARIO LAND REGISTRATION REPORT, supra note 189, at 99-100. On the ba-
sis of available data, the Commission assumed that searches averaged 167 minutes in
recording and 66 minutes in registration. The Commission assumed that the search-
ing would (or could) be done by paralegals (conveyancers). The assumed average
title examination time by lawyers was five minutes longer for recording than for re-
gistration. Risk estimated the additional time to be 5 to 10 minutes. Id. at 100.
Some American recording offices probably are not as well organized or managed as
the Ontario offices studied by the Commission and, consequently, searches in those
American offices are probably more time consuming.
211. Stanton S. Roller, Title Insurance-An International Perspective, TITLE NEWS,
Mar. 1983, at 7.
1992] McCormack: Torrens TORRENS
and Recording:
ANDLandRECORDING
the paucity of reliable and meaningful comparative user cost

data, published reliable data on comparative governmental
costs are available. These data include information from the
County of York, Ontario, and three American jurisdictions
with Torrens and recording systems operating side by side: Il-
linois (Cook County), Minnesota (Hennepin County) and Ohio
(Hamilton County).
Comparative governmental costs have frequently been over-
looked in published commentary comparing Torrens and re-
cording and are of vital significance in explaining the failure of
Torrens to take hold in much of the United States. To work
efficiently and satisfactorily, a title registration office must be
staffed adequately with competent people who are capable of
swiftly and accurately evaluating instruments presented for re-
gistration. These staffing requirements make it more costly to
run a title registration office than a recorder's office.
In 1978, under contract from HUD, Booz, Allen and Hamil-
ton, Inc., conducted extensive field research on actual land rec-
ord-keeping activities at twenty sites in the United States.2 12
The Booz, Allen report on this research contains comparative
cost data on three locations with dual Torrens and recording
systems: Cook County, Illinois; Hennepin County, Minnesota;
and Hamilton County, Ohio.
The following table is excerpted from a table in the Booz,
21 3
Allen report:
2 14
RECORDER'S COSTS: COOK HENNEPIN HAMILTON
Documents Processed 246,983 88,060 82,584

Transfers 81,112 26,683 27,230
Total Expenditure $1,556,628 $536,964 $425,902
Cost Per Document $6.30 $6.10 $5.16
Cost Per Transfer $19.19 $20.12 $15.64
212. HUD, STATE-OF-THE-ART STUDY, supra note 18, at 68.

213. Id. at 71. Anyone who evaluates these data should be aware that the three
Torrens systems were hardly equivalent. For example, Cook County checked the
signatures on documents presented for registration for forgeries; Hennepin County
did not. Also, at the time, Cook County government was still saddled with the bur-
dens of a now largely defunct political patronage system. When recording and Tor-
rens costs are compared, it should be appreciated that the Registrar or Examiner of
Titles performs quasi-judicial functions. The Registrar or Examiner determines title
questions which otherwise might have to be handled by the general court system.
Edblom Letter, supra note 101, at 2. There are no available data which could be used
to assess the economic impact of these quasi-judicial functions.
214. Excludes costs incurred by Chicago Title Insurance Company.

2 15
TORRENS COSTS: COOK HENNEPIN HAMILTON
Documents Processed 35,825 50,310 15,403

Transfers 16,324 17,000 5,080
Total Expenditure $2,000,084 $546,000 $104,604
Cost Per Document $55.83 $10.85 $6.79
Cost Per Transfer $122.52 $32.12 $20.59
The Ontario Law Reform Commission data for the County

of York also indicate that the cost to government of operating
a title registration office is greater than the cost of running a
recording office. The cost of staff for each recordation in the
recording office was $4.50, the cost of space was $2.50 and the
entire cost for one recordation was $7.75. To register a title,
the cost was higher. The cost of staff for a registration was
$6.50, the cost of space was $2.25 and the entire cost for a
registration was $9.50.216
The Booz, Allen study also reveals that American recorders'
offices are profit centers for counties.2 1 7 In contrast, American
Torrens offices usually require a subsidy because the income
they receive falls short of operational costs. 21 8 The high cost
of initial registration and the opposition from the title assur-
ance industry may have contributed to the failure of Torrens in
the United States. However, the high public cost and burden
of continuing administration of Torrens has been the most sig-
nificant contributing cause of its failure. For users, Torrens is
actually or potentially a more efficient, less costly system of
land title assurance and transfer than recording. Arguably, ti-
tle risks are or can be made to be so much less under Torrens
compared to recording that title insurance may be unneces-
215. Excludes costs incurred by Chicago Title Insurance Company.

216. ONTARIO LAND REGISTRATION REPORT, supra note 189, at 22.
217. In 1972, total operating expenses of U.S. recording offices were in excess of
$137 million; revenue received in the same year exceeded $240 million. HUD,
STATE-OF-THE-ART STUDY, supra note 18, at 13.
218. SHICK & PLOTKIN, supra note 37, at 58. "While there are considerable vari-
ances among the data-reporting formats used by the different systems examined, we
estimate that the 1976 'subsidy' ranged from $140,000 in Ramsey County to a high of
$625,000 for the statewide Massachusetts program." Id. Shick and Plotkin's cost
conclusions are in accord with recent Cook County data. The Blue Ribbon Commit-
tee appointed by the Cook County Recorder/Registrar estimated 1989 Torrens office
revenues of $3,305,000 and expenditures of $3,850,000. This deficit was expected
even though fully one-half of Torrens transfers are handled by deputized title insur-
ance company employees who are not on the County payroll. Cook County Blue
Ribbon Report, supra note 10, at 44. The Hennepin County Torrens office currently
requires no subsidy. Edblom Letter, supra note 101, at 2.
sary. But it is clear that Torrens or any other true registration

system is more costly and difficult for government to adminis-
ter than recording. This higher cost is inherent in title regis-
tration systems because much of the data consolidation,
evaluation and management done by private parties under re-
cording is done by government employees or their agents
under registration.2 9 This aspect of title registration cannot
be reduced to insignificance by reforms or improvements of
title registration procedures except by a radical and probably
unacceptable simplification of a jurisdiction's real property law
or a limitation of registrable interests to a very few types.
Since local governments had a first-hand appreciation of the
administrative difficulties and costs involved with Torrens, it is
not surprising that some of the most effective opposition to
title registration came from local governments responsible for
its implementation and not, contrary to conventional wisdom,
from the title assurance industry. 2 20 The worst enemies of
219. Experience in Ontario indicates that computerization can reduce this burden
on government.
220. The Cook County Recorder/Registrar led a successful effort to eliminate her
Torrens system, the oldest and one of the largest in numbers of titles registered in
the United States. No credible evidence exists that the title industry was behind this
effort to abolish Torrens. The industry certainly had little or no financial interest in
the outcome because it was insuring over 90% of Torrens transactions in Cook
County at the time the effort to eliminate Torrens was initiated. Title insurance com-
pany officials did serve on the committe which recommended abolition of Torrens.
Cook County Blue Ribbon Report, supra note 10, at 52.
The failure of Torrens in Cook County, after 90 years of implementation, was
mainly caused by incompetent, unsatisfactory administration. See supra note 11. The
demise of Torrens in southern California in 1955, after 40 years of implementation,
appears to have been due to grossly incompetent management. SHICK & PLOTKIN,
supra note 37, at 149. To attribute this incompetence to a title assurance industry
conspiracy to strangle Torrens is absurd. The "reputation of American county gov-
ernment for being incompetent," referred to by Professor Johnstone, is well
founded. Quintin Johnstone, Systems of Land Title Protection in the United States and Pos-
sibilitiesfor Their Improvement, reprinted in JAMES L. WINOKUR, AMERICAN PROPERTY LAW:
CASES, HISTORY, POLICY AND PRACTICE 1096 (1982). This incompetence is largely
due to the fiscal or political constraints which bind many county governments and is
really not the fault of the people who administer and work for these governments.
In some states, Torrens acts were repealed, not at the urging of the title indus-
try, but at the request of local recorders who found the existence of little used regis-
tration statutes to be a nuisance. SHICK & PLOTKIN, supra note 37, at 18; Charles D.
Knight, Comment, A Fly-Specker's Manualfor the Illinois Torrens Act, 1978 U. ILL. L.F.
487, 488 n. 13. In Minnesota, Torrens administrators successfully lobbied for amend-
ments making possessory registration available at the option of each county only
where a part of the applicant's tract was already registered. The administrators
feared the costs and burdens that a relatively sudden and substantial increase in re-

Torrens included administrators charged with implementing it

who were generally ill-equipped and disinclined to do so. In-
formed 22 1 government administrators are not likely to wel-
come with enthusiasm a system that is more difficult and costly
for them to administer. Informed local legislators and admin-
istrators normally will not support replacing a system that gen-
erates profits with one that may require a subsidy. The high
cost to local government for continued administration tends to
make government unsupportive of registration, unless ade-
quate resources are allocated to support the greater burden.2 2 2
The failure to provide adequate resources has contributed to
an inertia that caused little registration to occur after twenty
gistrations would place on county government. Comment, Possessory Title Registration,

The conventional wisdom that the title assurance industry was a major cause of
the failure of Torrens in the United States is simplistic. As Professor Percy Bordwell
pointed out in 1940, the conventional wisdom does not explain why Torrens was a
relative success in the state legislatures (20 states or territories enacted Torrens acts
between 1895 and 1917, when both state legislatures and the title assurance industry
were dominated by lawyers) and an almost complete failure in implementation. If the
title assurance industry was a major force in blocking Torrens, one wonders why the
industry was relatively ineffectual at the state legislative level. See Percy Bordwell, The
Resurrection of Registration of Title, 7 U. Cm. L. REV. 470 (1940). The title assurance
industry may have contributed to the failure of Torrens by remedying problems
caused by weaknesses in the recording system and thereby removing the need for its
replacement.
The notion that the title assurance industry's lobbying against or disparagement
of Torrens significantly caused its failure is not based on established fact. Today,
correctly or not, title insurers generally believe that Torrens is cumbersome, archaic
and no serious threat to their business. For example, William McAuliffe, the Senior
Vice President of the American Land Title Association, stated that the Torrens sys-
tem "has not proven itself" and is unlikely to " 'put the title companies out of busi-
ness.' " Cheryl Frank, Title Firms Dying?, A.B.A. J., Nov. 1985, at 33. Furthermore,
according to Leonard Donahoe, General Counsel for Chicago Title Insurance Com-
pany, even in an ideal system, " '[t]here are still going to be elements of risk, expo-
sure and potential loss.' " Id. Thus, property owners may still need title insurance.
221. A source of information about the costs and administrative difficulties in-
volved with Torrens was almost certainly the title assurance industry.
222. However, consider a 1972 newsletter issued by the then Cook County Re-
corder/Registrar, the late Sidney Olsen, who touted the alleged superiority of regis-
tration and urged owners to register. See Sidney R. Olsen, A Dozen Reasons Why You
Should Register Your Property Under the Torrens System, COOK Co. RECORDER, Sept. 1,
1972, at 9. Despite his urging. few Cook County real estate owners followed his
advice. Between 1969 and K-31, only 120 initial registrations occurred. Cook
County Blue Ribbon Report, supra note 10, at 40. No initial registrations have been
made in Cook County since 1981. Id. The current Cook County Recorder/Registrar,
Carol Moseley Braun, a lawyer and former state legislator, led the successful move-
ment to abolish Torrens in Illinois. Recorder Urges End to Torrens System, supra note
195, at 61.
states passed enabling legislation, except in a few instances

where special facts and circumstances overcame this inertia. 22
IV. COMPUTERIZED RECORDING
A. Registration and Recording: Land Title Assurance for the

Computer Age
A world-wide movement is advancing toward computerized

22 4
land title record systems. American title insurance compa-
nies have been developing computerized land title record sys-
tems since the 1960s.225 American local governments have
been implementing unofficial computerized land title record
systems since the late 1970s. In Australia, computerization of
a Torrens system began in 1969.26 By 1982, computerization 227
of the Austrian title registration system was well underway.
In 1984, Ontario, Canada, authorized an official computerized
recording and title registration system.228 This movement will
accelerate as less expensive, more powerful and more user-
friendly computer technology develops. Inevitably, computer
systems will replace manual records, especially in populous ju-
risdictions with frequent land transfers and large amounts of
data to manage. 29
223. In Cook County, these special facts included the need for reestablishing title
where the 1871 Chicago fire destroyed the public land title records.
224. See Lang, supra note 35, at 213. Bumps in the road to computerization have
been encountered. The effort to computerize the Hennepin County, Minnesota,
Torrens office was a near disaster. The failure of the Hennepin County project was
the result of the decision to use computers to produce paper certificates. Experience
with computerization of title registration in Ontario shows that computers are more
effectively used when paper records are completely replaced with electronic storage.
The computer, of course, can print paper copies when needed. Edblom-Witkowski
Interview, supra note 72; John Dalgliesh Interview, supra note 45. See also David L.
Drury, Computerization Increases Productivity, TITLE NEWS, Sept.-Oct. 1986, at 13 (stating
that real estate recovery and 100% increase in volume brought increasing pressure to
computerize). Computerization proved to be no panacea for the Cook County Re-
corder/Registrar. Mark Hornung, Best-laid Plans FallShort in Recorder's Office Revamp,
CRAIN'S CHI. Bus., Mar. 5, 1990, at 56. The author cynically commented: "But only
in Cook County government could computerization make matters worse." Id.
225. Fiflis, Security and Economy, supra note 2, at 173.
226. See Lang, supra note 33, at 213.
227. Id.
228. Land Registration Reform Act, R.S.O., ch. 32 (1984) (Can.).
229. Manual records may be more cost effective in smaller systems with less data
to manage. Cf Drury, supra note 225, at 13.

B. Basic Components
A computerized land title assurance system, whether record-
ing or registration, should have two basic components. First, it
should contain a database with information on ownership and
title, consisting primarily of the facts pertaining to past title
transactions and past legal evaluations of these facts. Second,
it should contain maps showing the units of ownership in the
system with permanent identifying symbols assigned to each
unit. The database should be organized into computer records
for each ownership unit in the system shown on the map. This
proposed system is essentially a computerized version of the
Austrian, German and Swiss title registration systems. 23 0 The
Chicago Title Insurance Company uses a similar plan for its
computerized title plant in Cook County, Illinois.231 It also is
230. DowsoN & SHEPPARD, supra note 30, at 173-90.

231. In Cook County, both the Chicago Title Insurance Company and the County
maintain computerized land title records. Access to data in both systems is through
permanent identification numbers assigned to parcels. Interview with Sally Dolphin,
Chicago Title Insurance Company, in Chicago, Illinois (April 1989) [hereinafter
Dolphin Interview]. The Cook County system of permanent identifiers was devel-
oped by Mr. Joseph Sidwell who owned a mapping company in the 1940s. Interview
with Robert Egan, Office of the Assessor, Cook County, in Chicago, Illinois (April
1989). Sidwell mapped the entire County into townships, sections, blocks and par-
cels. Each parcel was assigned a permanent identification number. Sidwell per-
suaded Cook County to use his system and maps to identify parcels for real property
taxation purposes. The Illinois legislature authorized the use of Sidwell's system.by
Cook County in 1945. See ILL. ANN. STAT. ch. 120 511 (Smith-Hurd 1969 & Supp.
1991) (current version of the authorizing legislation). The maps are updated as
changes occur. These maps resemble the governmentally maintained cadastral
surveys found in continental Europe and other parts of the world. See DowsoN &
SHEPPARD, supra note 30, at 46-71. The POLARIS maps in Ontario are similar, but
they are computer-based; hard copies are printed for the convenience of users. Dal-
gliesh Interview, supra note 45.
In Cook County, the County Clerk is responsible for assigning identifiers to spe-
cific parcels. The assigned number refers to a particular parcel in a particular loca-
tion but does not determine precise boundaries. Precise legal descriptions can be
obtained from Sidwell maps or other sources. Subdivisions and consolidations are
handled administratively. When new numbers are issued, the former ones are per-
manently retired.
The Sidwell map system existed for decades before computerization of land title
records began in Cook County. Because of the success of the Sidwell system over
decades and its adaptability for computer data retrieval use, both Cook County and
Chicago Title Insurance Company are presently using it to identify and retrieve data
from their computer systems. Since 1986, the Cook County Recorder/Registrar has
required instruments presented for recordation or registration under Torrens to
bear the permanent identification numbers assigned to the parcels affected by such
instruments.
Prior to computerization, the Chicago Title and Trust Company maintained five
1992]
TORRENS AND RECORDING
the basic system currently being implemented in Ontario, Can-

ada, for its title registration system.23 2
Data pertaining to each title would be stored in a computer
record permanently identified by a number, called a land par-
cel identifier. Individuals who are parties to documents can be
identified by name and birth date; 233 other entities can be iden-
tified by name and taxpayer identification numbers. Each file
would contain a list of the documents entered. Photographic,
electronically stored copies of title transaction documents
would be retrievable upon command.23 4 The party initiating
registration or recording should be required to supply the land
parcel identifiers. This requirement will make the indexing
even less burdensome for government than administering a
grantor-grantee index under a recording system. When a
transfer of ownership is recorded or registered, the transfer
would be added to a file which will be similar to a parcel-spe-
cific tract index. While the land parcel identifiers should be
manual indexes to its elaborate Cook County title plant: tract, general tax, special
assessment, judgment and miscellaneous (bankrupts', minors' and decedents' es-
tates) indexes. Fiflis, Security and Economy, supra note 2, at 173. By 1965, all but the
tract index had been computerized and put into at least partial operation. Id. Com-
puterization of the tract index was completed in 1974. Dolphin Interview, supra.
For convenience, in the Chicago Title Insurance Company system, land title data
in its title plant may be accessed through either permanent identification numbers or
street addresses. If the permanent identification number is unknown, entry of the
street address will retrieve the permanent identification number and other data perti-
nent to the parcel. When the company has previously searched or examined the title
to a parcel, the results are entered in the database. When the title is searched again,
these results will be retrieved. The company will normally only search back to the
date of the prior examination. This procedure supplies the "curtain" feature found
in title registration systems which is intended to eliminate the need for lengthy his-
torical searches.
Legal description and permanent identification numbers are matched and
checked using Sidwell maps. The Sidwell maps contain detailed legal descriptions
with permanent index numbers assigned to all parcels shown on the maps. If the
legal description of a parcel is known, the permanent identification number for it may
be retrieved from the Sidwell maps and vice versa.
232. Dalgliesh Interview, supra note 45; Dow, supra note 45, 9.06[l][d]; Lamont,
233. Federal privacy legislation may preclude use of social security numbers and,
in any event, persons may refuse to disclose them. Birth dates, however, are matters
of public record in the United States.
234. One of the very few weaknesses of the Ontario POLARIS system currently
being implemented is that data from registered instruments is manually typed into
the database. Human typists, unfortunately, make mistakes. However, so do elec-
tronic scanners. As computer technology advances, documents will be photographi-
cally stored and retrieved by letters, words or symbols contained in them.

WilliamWILLIAM
Mitchell LawMITCHELL
Review, Vol. 18,
LAW Iss. 1REVIEW
[1992], Art. 4 [Vol. 18
the primary key to accessing file information, this information

should also be retrievable by persons' names, street addresses,
document numbers assigned to recorded documents and other
useful means.
Identifying land titles and parcels by permanent land parcel
identifiers is not as simple as identifying humans with perma-
nent numbers such as social security numbers. Humans can-
not be subdivided or consolidated. The development of a
workable system of permanent land parcel identifiers is not a
simple undertaking, as dimensions of parcels may change
through subdivisions, consolidations, and other accretions and
deletions.2 3 5 A good system of permanent identification num-
bers also requires accurate and current maps showing the par-
cels identified.23 6
During the drafting of the Uniform Simplification of Land
Transfers Act (USOLTA), the permanent identification
number approach was rejected.2 3 7 Instead, the drafters pro-
posed a system where the recording jurisdiction would be di-
vided into segments such as squares which would be assigned
locator numbers. Instruments would be indexed according to
the locator number assigned to the square or other segment in
which the subject property was located. Under this proposal,
the title searcher would sort out the data applicable to the title
being examined from data pertaining to other properties in the
same recording district with the same identifier. According to
one of the drafters, Professor Dunham, their proposal of per-
manent identification numbers avoids the problems of consoli-
dations, subdivisions and the need for accurate maps. 3 8
235. See Allison Dunham, Land ParcelIdentifiers and the Uniform Land TransactionsAct,
43 U. CIN. L. REV. 469, 475-86 (1974).
236. Id.
237. Id. at 480. The indexing section is in the Uniform Simplification of Land
Transfers Act which was spun off the Uniform Land Transactions Act. See UNIF. SIM-
PLIFICATION OF LAND TRANSFERS ACT 6-207, 14 U.L.A. 368 (1990).
238. Professor Dunham also states that "assignment of a separate number, even if
geo-coded to the center of the parcel, does not aid appreciably in performance of
another role of the conveyancer-examining descriptions of neighboring parcels and
the title history of these parcels to determine whether there are orts, gores, and over-
laps and also rights in land of another." Dunham, supra note 236, at 481. Under a
permanent number system, the title searcher can easily retrieve data about adjacent
parcels through the permanent numbers assigned them as shown on the map. Under
the drafters' system, only data for other parcels in the same square will be retrieved.
If the subject property is on the edge of the square, information about adjacent prop-
erty will not be seen unless data for the adjacent square is also retrieved. Id. at 482.
However, their proposed system would be more cumbersome

to use, especially in densely developed areas with many titles in
small areas.23 9 Consolidations and subdivisions under perma-
nent number systems can be handled administratively without
great difficulty. 24 0 Consolidations involving entire parcels do
not absolutely require the issuance of new permanent num-
bers; the numbers assigned the former separate parcels can be
used to describe the consolidation. Subdivisions present more
difficulty because they result in two or more separate parcels.
In that case, a new identifier must be assigned to each parcel.
However, this can be done by adding numbers for each new
segment to the original identifier for the subdivided parcel.
The need for maps for permanent number systems is a more
serious, though surmountable, obstacle. Even under the
USOLTA drafters' proposal, a map would be needed to locate
a specific property within the proper square or squares.24 ' Ac-
cording to Dunham, the "searcher or the person initiating re-
cording can enter the locator into as many locator squares as
he thinks the parcel falls .. . If the searcher and the per-
son initiating recording "think" differently about which locator
squares should properly be entered, a problem may arise.
Maps supporting permanent number systems do not have to
be perfect. The depiction of the boundaries is sufficient if the
map adequately identifies the various numbered parcels so that
each can be distinguished from other parcels. Therefore, a
239, Professor Dunham refers to a hypothetical square of land in the Hyde Park
area of Chicago with "a portion of the University of Chicago and portions of 2 or 3
faculty houses" and states that this square "may have only 3 or 4 entries a year speed-
ily retrievable" under the proposed system. Id. at 483. The author is familiar with
other areas of Chicago where the proposed system would not work so well. In the
lake shore area of Chicago, there are blocks containing many high rise condomini-
ums containing hundreds of units and, therefore, separate titles for which there
would be many more annual entries indexed for a recording district square of land
the approximate size of Dunham's hypothetical square near the University in Hyde
Park.
240. In Cook County, subdivisions and consolidations are handled administra-
tively. A petition for Divisions or Consolidations is filed with the Division Depart-
ment of the County Assessor's Office. These petitions currently take about two years
to process because of the number of petitions received. The process could be accel-
erated considerably with better management and not a great deal more resources.
The Ontario POLARIS system uses a similar procedure when new numbers are
needed. Dalgliesh Interview, supra note 45.
241. See UNIF. SIMPLIFICATION OF LAND TRANSFERS ACT 6-206, 14 U.L.A. 368
(1990).
242. Dunham, supra note 236, at 482.

WilliamWILLIAM
Review, Vol. 18,
LAIss.
W 1REVIEW
[1992], Art. 4 [Vol. 18
costly, comprehensive survey is not necessary. Adequate maps

for zoning ordinances are assembled without costly compre-
hensive surveying. The maps can be assembled from existing
tax assessment maps or subdivision and condominium plats.2 4 s
Maps in many rural areas could be derived largely from the
U.S. Government Rectangular Survey System and aerial map-
ping where necessary. Digitizing 24 4 the boundaries of individ-
ual parcels on computers is a cost-effective way of assembling
computer-based maps.2 4 5
All in all, the permanent identification number approach
would appear to better promote economy, efficiency and se-
curity of title.2 4 6 The USOLTA drafters' proposal appears too
cumbersome to use with few compensating advantages to
counterbalance its faults.
The final Official Draft of USOLTA requires "recording of-
ficers" to maintain geographic indexes and "maps that indicate
location in a manner enabling public users to find the proper
location or locations in the geographic index for every land
parcel. ' 24 7 The definition of "geographic index" is broad
enough to include a permanent identification number system,
the system described by Professor Dunham, or almost any con-
ceivable tract index, whether it is manual or computer-
based.2 4 8
All system users should pay fees to access the data absent a
showing of financial hardship. Current land title records sys-
243. Such maps are the initial source material for the computer-based maps being
developed for the Ontario POLARIS system. Information from survey plans and
registered title documents are used to build an accurate mapping database. Govern-
ment does not affirm the boundaries shown by the POLARIS computer-based maps.
Dalgliesh Interview, supra note 45.
244. Digitizing is the computerized process of recording of numerical representa-
tions of geographical points on a map. Lang, supra note 35, at 214. "Graphic record-
ing, or digitizing, involves the recording of individual land parcels, by manually
pointing a device (like an electronic pen) at the lot corners already drawn on the
standard maps. The computer then automatically calculates the lot's position on the
earth's surface." Id.
245. See Lang, supra note 35, at 214.
246. Ontario, Canada, adopted the permanent identification number method
when it decided to computerize its recording and title registration systems. See RE-
GISTRATION DIVISION, MINISTRY OF CONSUMER & COMMERCIAL RELATIONS, ONTARIO
DOCUMENT USER GUIDE, LAND REGISTRATION REFORM ACT, 1984 33,405 (1986 ed.).
247. UNIF. SIMPLIFICATION OF LAND TRANSFERS ACT 6-206 to 6-207, 14 U.L.A.
367-68 (1990).
248. Id. at 6-102.
1992] McCormack: Torrens and Recording:
TORRENS ANDLandRECORDING
tems are financed by fees from those who record or register or

are paid out of other government revenues.
C. Registration or Recording?
Computerization has fundamentally changed the debate be-
tween the Torrens and recording systems. Computerization of
recording can eliminate redundancies and eliminate many
other deficiencies of the recording system by providing rapid
access to relevant data and eliminating redundancies. Further,
remote terminal access to computerized public records will re-
move most of the incentive for title companies and examiners
to maintain private records.2 4 9 By contrast, Torrens already
has good data management and retrieval characteristics, even
without computerization.
The most difficult and important policy question faced by
architects of computerized land title assurance systems is
whether they should be purely recording systems or at least
partially title registration systems. Persons considering this
question should keep in mind that the vital distinction between
the two systems lies in "the affirmations made by the state
about the ownership of interests and the effect of docu-
ments. "250 Fully computerized recording and title registration
systems should be very similar both in physical characteristics
and in the manner by which data are stored, managed and re-
trieved. The basic issue is the extent to which the government
will affirm the legal effect of information retrieved from the
systems.
The best system would provide great security of title with
low governmental and user costs. For most of the United
States, the appropriate choice would be a computerized re-
cording system, with privately or publicly supplied title insur-
ance, which effectively incorporates the mirror, curtain and
indemnity principles of title registration. 25 ' Experience with
title registration in the United States and elsewhere in com-
249. Title service providers sometimes use their private title plants to maintain a
competitive advantage over others who are forced to use poor public records. Good
public records will remove this advantage. Ontario is considering remote access to
POLARIS from terminals in solicitors' offices. Dalgliesh Interview, supra note 45.
250. Risk, supra note 5, at 471.
251. Of course, title registration may be appropriate in those few American juris-
dictions where Torrens has worked well, such as in Hennepin County, Minnesota.
Title registration can succeed in jurisdictions which have skilled and dedicated peo-

William WILLIAM
Review, Vol. 18,
LAWIss. 1REVIEW
[1992], Art. 4 [Vol. 18
mon law countries indicates that conclusiveness of the register

is an elusive, often unattainable goal.2 5 2 Nonetheless, existing
title registration systems are less costly for title searchers and
examiners to use and probably provide greater security of title
compared to existing recording systems. However, the price
for these user benefits is greater governmental cost and admin-
istrative difficulty as well as costs and difficulties encountered
by system users. A properly structured computerized record-
ing system which effectively incorporates the title registration
principles of mirror, curtain and indemnity may provide equal
or greater economy, efficiency and security of title.2 53
For countries such as the United States, which have en-
trenched recording systems and little experience with title re-
gistration, recording has many attractive features unrelated to
its inherent merit. A reformed, computerized recording sys-
tem would be far easier to implement than a computerized re-
gistration system. For example, no costly judicial or
administrative initial registration proceedings would be
needed with official computerized recording. Implementation
of a computerized title registration system would face the same
barriers of high start-up costs due to initial registration, and
resistance due to administrative difficulties and costs which
caused most manual Torrens systems to fail in the United
States. Computerization will do nothing to remove these bar-
riers. In addition, relatively few American lawyers and judges
understand title registration. Since recording is much better
understood, educating lawyers and judges about a reformed,
pie available to operate the systems well, and a public and governmental commit-
ment to raise and expend the funds needed for efficient operation.
252. MAPP, supra note 5, at 181; Rose, supra note 114, at 588-90.
253. A Canadian study committee also proposed a merger of recording and title
registration concepts. After recognizing that it is not possible for any title registra-
tion system to guarantee ownership of all interests in land and that recording is more
efficient and user friendly at the stage of data input than is title registration, the com-
mittee proposed a system where only the most common and best understood inter-
ests would be registered (fees simple absolute, life estates, leaseholds, security
interests, easements, utility interests and restrictive covenants); all others would be
recorded. JOINT LAND TITLES COMMITrEE, RENOVATING THE FOUNDATION: PROPOSALS
FOR A MODEL LAND RECORDING AND REGISTRATION ACT FOR THE PROVINCES AND TER-
RITORIES OF CANADA 14, 21 (1990). This proposal merits serious consideration.
However, it may be more appropriate for Canada than the United States because
Canada has had much more experience with title registration and does not have the
legal, political and practical barriers to its implementation to the extent found in the
United States.
1992]
McCormack: Torrens TORRENS
AND RECORDING
official computerized recording system would be less

formidable.
The most obvious inherent advantage of recording is that it
is less costly and less difficult for government to administer.
Staffing a recorder's office is relatively easy because little ex-
pertise in real estate law is required. In contrast, at least some
employees in a title registration office must have substantial
expertise. Furthermore, because recording places relatively
few demands on government resources, it is more responsive
to sudden increases in the volume of data input than is title
registration. Also, few resources are actually expended to
close transactions in recording because government does not
substantively review documents before recording.254 More-
over, no time is wasted because of adjourned closings due to
governmental objections.
The advantages, however, bring on their own disadvantages.
Lack of examination means that the recording system provides
no formal control on the quality of data input. Title insurance
companies, lenders, escrow agents or lawyers representing
parties may indirectly contribute such control. Also, a non-
binding quality check for errors at the point of data input,
which is similar to procedures used in registration offices,
might be a cost-effective way of promoting security of titles and
efficiency. This quality check would avoid creating the inflexi-
bility caused by a binding, conclusive approval procedure.255
Conveyancing on state-prescribed forms whenever possible
would elevate the quality of data input and promote economy
and efficiency.
D. Mirror, Curtain and Indemnity in Computerized Recording

Systems
Architects of computer-based recording systems should
draw on decades of experience with title registration systems
in the United States and elsewhere. The title registration prin-
ciples of mirror, curtain and indemnity should be applied,
254. However, the actual closing costs may be high if fees and charges are
excessive.
255. A non-binding review should be less prone to the overly meticulous examina-
tion encountered in binding reviews in some title registration offices. In contrast to
American practice, some European recording systems are quite restrictive about ac-
cepting instruments presented for recordation. See GARRO, supra note 5, at 111-14.

William WILLIAM
Review, Vol. 18,
LAWIss. 1REVIEW
[1992], Art. 4 [Vol. 18
whenever possible, consistent with the goals of economy, effi-

ciency and security of titles.
1. Mirror
The mirror principle of title registration should be a para-
mount policy. Generally, all liens, encumbrances and other in-
terests must be recorded in the computer file to give record
notice to subsequent purchasers.
The real estate tax lien or its equivalent is the only practi-
cally universal overriding interest in title registration systems.
A requirement to file these liens periodically would be too
great a burden on government, especially since general knowl-
edge of the liens minimizes their impact on the conclusiveness
of the register. While no great harm would result in exempt-
ing these liens from recordation in the computer files, com-
puterization would make their recordation a simple process.
Computers can be programmed to make automatic entries
showing the attachment of annual real estate tax liens against
non-exempt property. Special assessments do not attach at
regular intervals against titles generally, so they should not be
exempt from recordation in the computer files. Rather, special
assessments should not attach to titles until entered in the par-
cel files for the properties affected. Entries showing payment
or release of both general real estate tax liens and special as-
sessments could be made through a direct connection to the
office which receives payments.
As required by some Torrens acts, judgment liens should be
recorded against individual parcels. Recording and indexing
such liens in a general judgment lien index should not suffice.
During the drafting of the USOLTA, the drafters were per-
suaded by representatives of creditors' interests who sug-
gested that it would be politically impossible to eliminate
general lien indexes.256 Notwithstanding this alleged "political
impossibility," in both Minnesota and Illinois, state statutes re-
quire that judgment liens against Torrens titles be filed against
individual Torrens certificates in order to be perfected.2 5 7
256. Peter B. Maggs, Land Records of the Uniform Simplification of Land Transfers Act,
1981 S. ILL. U. L.J. 491, 502.
257. ILL. ANN. STAT. ch. 30, 122 (Smith-Hurd 1960 & Supp. 1991); MINN. STAT.
508.63 (1990); United States v. Ryan, 124 F. Supp. 1, 6 (D. Minn. 1954). In Con-
necticut, judgment liens must be filed against specific parcels described by the lien
claimants. CONN. GEN. STAT. ANN. 52-380(a) (West 1991). The drafters of
Transfers by will or inheritance should be recorded in the

appropriate computer file. Adverse possession should be pro-
hibited, as is the case under Torrens acts. At the very least,
those claiming title by adverse possession should be required
to file notice of their 8claims in the appropriate parcel file to
25
perfect their priority.
The hidden mechanic's lien should be abolished. The
USOLTA proposals for construction liens provide an excellent
guide for reform. Notice of such liens should be filed by parcel
before work commences and their general priority should re-
late to the time of filing.
As is the case with some existing Torrens systems, there
should be no inquiry notice or constructive notice from pos-
5 Interests not entered in the computer files for the
session. 259
subject titles should be deemed unrecorded. However, leases
should be exempt from recordation if they are short-term
(three years or less) and the lessees are in possession of the
property.
2. Curtain
Incorporating the curtain feature of title registration into a
computerized recording system is a difficult problem. The cur-
tain of existing title registration systems is based on an attempt
to make the register a generally conclusive governmental state-
ment of the condition of title, the very aspect of Torrens which
caused it to fail in most of the United States. This characteris-
tic is responsible for the high cost of initial registration and
continuing administration, the cumbersome features and in-
flexibilities of Torrens, and the inefficiency of continuous data
evaluation as instruments are presented for registration. Fur-
thermore, conclusiveness of the register may be unattaina-
ble. 260 As is the case under existing recording systems, the
record under computerized recording systems can be made
USOLTA should have left it to state legislators to determine whether elimination of

general lien indexes would be "politically impossible."
258. See Whitman, supra note 15, at 57.
259. See PA'rON & PATrON, supra note 20, at 682; Taylor Mattis, Recording Acts:
Anachronistic Reliance, 25 REAL PROP. PROB. & TR.J. 17, 100 (1990); see also Rosewood
Corp. v. Illinois Bell Telephone Co., 230 N.E.2d 172, 173 (Ill. 1967).
260. MAPP, supra note 5; Rose, supra note 114, at 588-90.

WILLIAM MITCHELL IA W REVIEW [Vol. 18
presumptively valid 26 1 without all of the costs, delays, inflexi-

bilities and other difficulties caused by the attempt to make the
register conclusive.262
Without some kind of curtain, user data evaluation costs
under recording are unnecessarily high because identical data
often have to be reevaluated from time to time. Existing meth-
ods of avoiding reevaluation include marketable title acts, title
examination standards and reliance by title examiners on prior
examinations. Reliance on prior examinations would be more
effective to promote economy and efficiency if the process
were formalized and better coordinated. Informal curtains
may come from routinely entering title insurance company title
reports or lawyer title opinions for individual titles in the com-
puter files for the titles examined. The recording office could
make the title insurer's or other examiner's direct access to the
database through remote terminals conditional on their enter-
ing title reports or opinions in the database.263 Alternatively,
recording title opinions or reports could be required to qualify
transfer documents for recording. 2 4 Subsequent title examin-
ers could then choose to rely on prior reports or opinions in
determining the extent of their searches. At present, title ex-
aminers rely on past examinations in a sporadic, disorganized
fashion. Implementation of this proposal would organize and
routinize this practice. Where a title insurer relies on the cur-
tain of a prior title insurance company report, security of title
could be enhanced by making prior owners' title insurance pol-
icies assignable to present owners with expiration to occur af-
ter some period of time, such as thirty or forty years after
261. This is done through a presumption that recorded transactions were legally
effective.
262. See supra notes 252-56 and accompanying text.
263. Computer records organized by parcel will eventually eliminate the use of
abstracts of title. The title examiner will not have to rely on abstractors' summaries
because electronically stored copies of original documents will be almost instantly
retrievable from remote terminals.
264. Recording title opinions is not a new idea. See, e.g., Fairfax Leary, Jr. & David
G. Blake, Twentieth Century Real Estate Business and Eighteenth Century Recording, 22 AM.
U. L. REV. 275 (1972). The authors propose making title opinions conclusive after
they had been on file for a set period of time, such as six years. Id. at 318. This
process would create a private sector title registration procedure. Conclusiveness is
probably not necessary to provide the needed curtain; one wonders how quickly
courts and legislatures would heap "mud" on this crystalline proposal. See Rose,
1992]
AND RECORDING
issuance.2 6 5
Governmental curtains cutting off the relevance of past
transactions may be provided by an administrative certification
of title procedure similar to that provided by the Ontario Cer-
tification of Titles Act. 266 This Act provides for a type of ad-
ministrative quiet title procedure which can be used to
establish conclusively the state of a recorded title as of a partic-
ular time.
3. Indemnity
The indemnity feature of title registration also should be
adopted. At the very least, indemnification should be available
to compensate those who suffer losses because of mistakes, er-
rors or omissions in administration. The current procedure of
suing the recorder on his or her bond is woefully inadequate.
A claimant should not have to bring a lawsuit to establish a
claim. Experience with the administration of Torrens assur-
ance funds indicates that claim payment policies should be
fairly liberal in order to promote confidence in the funds.
Claimants should not be disqualified because they may be enti-
tled to sue others for their losses. In such cases, government
265. See Whitman, supra note 15, at 57. Loan policies insuring mortgagees are
currently assignable and have been for many years. Loan policies have built-in life-
times because they expire when the mortgages are discharged. Title insurance com-
panies might very well object to being required to issue policies which could
potentially insure a perpetual chain of owners even though the liability of the com-
pany under any assigned policy would be limited to covered losses in existence at the
date of the policy.
266. R.S.O. ch. 59 (1970); Richard E. Priddle, New Requirements and Procedures in
Land Titles and Registry Offices, in
SPECIAL LECTURES OF THE LAW SOCIETY OF UPPER
CANADA 1970, RECENT DEVELOPMENTS IN REAL ESTATE LAW 353, 356-57 (1970). This
Act engrafts onto the recording system the generally conclusive governmental state-
ment of ownership and title feature of title registration. Under this Act, which has
limited geographical application, the administrator of the land title registration sys-
tem has the authority, upon application, to have a certificate of title issued which
establishes ownership and title as of the time of issuance. Essentially, the certificate
has the effect of a final judgment of quiet title although it may be issued administra-
tively. After issuance of the certificate, the property remains in the recording system.
The advantage is that the legal effect of prior transactions is, in general, permanently
determined. According to one commentator, the Ontario Certification of Title Act is
infrequently used. Dow, supra note 43, at 9.06[1][b]. "Such legislation is not re-
sorted to often in Ontario, since . . . [land under recording] may voluntarily be
brought within the land titles [registration] system and all, rather than only some, of
the benefits of such system thereby obtained." Id. Another commentator indicates
that the Certification of Titles Act is more frequently used. See Lamont, supra note 5,
at 102.

should indemnify claimants in appropriate cases and be subro-

gated to their rights against others. Ordinary contributory
negligence should not bar claimants from recovery. If the in-
demnification component uses an assurance fund, the re-
sources of the fund should be backed up by government, as is
the case with some existing American Torrens systems. The
coverages currently available from title insurance companies
should be available from private companies or government.26 7
V. CONCLUSION
Previous proposals to reform land title assurance provide lit-

tle basis for an expectation that the reform proposals made in
this article will be promptly adopted and implemented. The
land law reformer should be sensitive to political realities.
Dreaming has merit, but progress ultimately depends on what
can be, not what should be. Nonetheless, meritorious ideas for
reform should not be held back because of expected political
opposition. If a sustained, serious effort is made to implement
a needed reform, special interest opponents will have difficulty
blocking it, so long as it appears sound and is understood by
legislators, other policy makers and the public.
An adversarial approach toward the title assurance industry
tends to ensure that potential allies will become enemies. It is
unwise to charge that people in the title industry are engaged
in an expendable, unnecessary activity to the detriment of the
public interest-especially because the charge is untrue. Peo-
ple in the industry have a formidable, detailed knowledge of
the workings of title assurance systems, including its public
and private components. 68 If approached as allies, people in
267. The Iowa Finance Authority has been authorized to issue what appears to be
a form of governmental title insurance coverage. See IOWA CODE ANN. 220.2,
220.91 (West 1985 & Supp. 1991). Iowa previously had been the only state which
prohibited title insurers from issuing policies within the state.
268. For example, one of the best general works on title insurance currently avail-
able was written by a title insurance company officer. See generally ROONEY, supra note
73. Rooney was the President of the Attorneys' Title Guaranty Fund, Inc., the bar-
related title insurance company in Illinois. The late Professor Robert Kratovil of
John Marshall Law School in Chicago was for many years a Vice President of Chicago
Title Insurance Company. Kratovil published extensively both before and after he
retired from Chicago Title and became a law teacher. He was a consultant to the
committee which developed the Uniform Land Transactions Act. See Michael T.
Rooney, In Memoriam: Robert Kratovil, 4 CONCEP'rs AND VIEWPOINTS 22 (1989).
the title assurance industry can be valuable sources of informa-

tion and creative ideas.
The governmental administrators who will implement re-
forms are usually very concerned about whether they will have
access to the money and other resources necessary to perform
the tasks assigned to them. A proposed system can only suc-
ceed if sufficient resources are provided to those who will have
to implement it. Fairly charging all system users, including
those who now have free access, would provide the needed
revenues.
Successful reform requires organization, leadership, time
and effort. Conscientious legislators will not adopt proposals
which may have substantial impact just because experts or au-
thority figures tell them they should. Legislators must be con-
vinced that the necessary effort and costs will be justified by
satisfactory results with no serious political costs. Because the
area of land title assurance is specialized and not easily under-
stood, the process of persuading and educating legislators will
require time, patience and a sustained effort.
A principal reason why the Uniform Commercial Code was
"the most spectacular success story in the history of American
law ' 2 6 9 is that its initiators and creators, under the leadership
of Professor Karl N. Llewellyn and Mr. William A. Schnader,
had the patience and the persistence to carry forward an effort
over fifteen years to persuade state legislators to adopt the
Code while continually making it more legally and politically
acceptable.270
The history of the Code's success provides valuable lessons
for anyone promoting legislative property law reforms in the
United States. While the Code's triumph is not a perfect
model for real estate law reform, it does teach us that property
law reform will succeed when the proposed legislation is well
drafted and considered, both legally and politically, and when
legislators are convinced that change is necessary.
269. JAMES J. WHITE & ROBERT S. SUMMERS, UNIFORM COMMERCIAL CODE 5 (3d
ed. 1988).
270. Id. at 4. In the beginning, their progress was glacial, but after over ten years
of sustained effort, the floodgates broke in the early 1960s when the Code became
the most successful piece of model or uniform legislation ever created in the United
States. Id. at 5.


Torrens and Recording - Land Title Assurance in The Computer Age

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Torrens and Recording - Land Title Assurance in The Computer Age

Hochgeladen von

Copyright:

Verfügbare Formate

William Mitchell Law Review

Volume 18 | Issue 1 Article 4

Torrens and Recording: Land Title Assurance in

Follow this and additional works at: h7p://open.mitchellhamline.edu/wmlr

TORRENS AND RECORDING:

t Associate Professor of Law, Loyola University Chicago School of Law; Mem-

Published by Mitchell Hamline Open Access, 1992 1

4. Possessory Interest Exceptions .................. 94

The weaknesses of the recording systems' used in the

proof that title registration could work in the United States.2

Published by Mitchell Hamline Open Access, 1992 3

data, the registration office will be unable to adequately admin-

through fees which produce adequate revenues. Edblom-Witkowski Interview, infra

Published by Mitchell Hamline Open Access, 1992 5

throughout the United States in the foreseeable future.' 2

by special land registration districts administered by qualified

II. TITLE ASSURANCE IN THE UNITED STATES

A. Development of Recording and Registration Systems in the United

Published by Mitchell Hamline Open Access, 1992 7

recording may raise a presumption of validity. 20 Furthermore,

20. 1 RUFFORD G. PATRON & CARROLL G. PATON, PATRON ON LAND TITLES 20

panies sometimes maintain unofficial tract indexes. Tract in-

Published by Mitchell Hamline Open Access, 1992 9

2. The Torrens System of Title Registration

the 1800s, England considered adopting a recording system

Published by Mitchell Hamline Open Access, 1992 11

enactment of the Land Registration Act of 1925. 34 In the late

was adopted by nineteen other states.4" Since then, the legisla-

3. Computerization of Title Records

Published by Mitchell Hamline Open Access, 1992 13

The Ontario system is now operational and in the process of

B. Title Assurance: Goals and Methods

49. See Whitman, supra note 15, at 41-42.

Published by Mitchell Hamline Open Access, 1992 15

as fully as possible with as little conflict between them as is

2. Methods of Title Assurance

tent, this inefficiency is reduced in some localities by reliance

an existing abstract. DEP'T OF HUD, LAND TITLE RECORDATION SYSTEMS: ELIMINAT-

Published by Mitchell Hamline Open Access, 1992 17

marketability by extinguishing certain old defects, their efficacy

insurance policy covers the insured even for losses caused by

Published by Mitchell Hamline Open Access, 1992 19

inefficient. As a result, some critics have proposed replace-

III. THE TORRENS TITLE REGISTRATION SYSTEM

A. The Torrens System in Operation: Mirror, Curtain and

the original. The owner's duplicate, or an acceptable substi-

Published by Mitchell Hamline Open Access, 1992 21

time thereafter, the owner or other party may be required to

a single claim. 75 Repeal of the California Torrens Act followed

Published by Mitchell Hamline Open Access, 1992 23

the required contribution to the assurance fund, the title must

82. Comment, Possessory Title Registration, supra note 78, at 838.

to reduce its cost by authorizing an administrative registration

Published by Mitchell Hamline Open Access, 1992 25

Presumably, these property owners perceived some advantage

C. Administrative Initial Registration of Possessory Titles

disincentive for Torrens registration-the high cost of regis-

96. Sclar, supra note 86, at 261.

Published by Mitchell Hamline Open Access, 1992 27

ular registration proceeding ... .'0

costs of identifying claimants to the title and giving them due

D. Torrens in Operation: Security of Title

108. As of November 26, 1991, 17 applications to register possessory titles in